196005

196005 196005

User Manual: 196005

Open the PDF directly: View PDF .
Page Count: 44

Download
Open PDF In Browser	View PDF

MAY

1960

•
VOL. 9 - NO. 5 & 58

Computer Programming for Command Control Systems
Using Computer Services in Small Business
News of Computers and Data Processors:
ACROSS THE EDITOR'S DESK

For engineering, for research, for busi ness
data processing in companies both large
and small ... The neVI, fully-transistorized

ELECTRONIC COMPUTING SYSTEM

~"

,:...,.,.,

+

another outstanding computer value
... marketed by Royal MCBee Corporation

Advanced design: fully transistorized-with important
new computer design concepts that provide the largest
memory, greatest problem-solving capacity and flexibility in the low- or medium-priced field. Entire system
-computer, input-output typewriter and tape punchread console-have been specially designed as a unit.
High-speed computing ability: extra large capacity
(8008 words) magnetic drum memory, with special
fast access features. Computing speeds of up to 230,000
operations per minute. Ultra high speed input-output:
500 characters per second photoelectric punched paper tape reader, and 300 characters per second paper tape
punch available as optional equipment. Easy to use:
maximum results can be obtained by non-technical
personnel. Users benefit from free training, continuing
assistance, an extensive library of programs. Versatile
command structure provides programming speed and
flexibility. Low in cost: priced just above the smallscale computers, the RPC-4000 outperforms computers
costing many times more. Economical to install and
2

operate: no site preparation, air-conditioning or special

maintenance required. Plugs into any standard wall
outlet. Multiple application ability: designed to perform engineering, scientific and research calculations,
as well as business data processing and management
control functions.
The RPC-4000 is a product of the Royal Precision
Corporation, and is marketed by the Data Processing
Division of Royal McBee. It is the latest member of the
growing family of electronic computers from the people whose LGP-30 has become the world's leading
small-scale computer.
•

Royal Precision Corporation

Royal Precision is jointly owned by the Royal McBee and General Precision Equipment Corporations. RPC-4000 sales and service are available coast-to-coast, in Canada and abroad through
Royal McBee Data Processing offices. For full specifications, write
ROYAL MOSEE CORPORATION, data processi ng division, Port Chester, N.Y.

COMPUTERS and AUTOMATION for May, 1960

*

The acceptance by business and
industry of the Philco 2000 All·
Transistor Data Processing System
has created a number of significant
advancement opportunities in our
organization both at our new head·
quarters in suburban Philadelphia
and at various key locations in
other parts of the nation. You are
invited to call, write or visit us to
discuss your future in our growth
organization.

COMPUTERS

an.d AUTOMATION

for May, 1960

3

COMPUTERS
and AUTOMATION
DATA PROCESSING
Volume 9
,Number 5

•

CYBERNETICS

ROBOTS
Established
September 1951

MAY, 1960

B~~J~~!:rl~r-JT~1;:'~s~1~~
Assistant
L:,'~' NE'IL', D.,

r

•

MACDONALlf

MosEs 'M;' BBRr:~N,

" PATIUC~

J.

'

'

Mc::GOVERN

":Edi(o{:~: :;

Assiitdnt: Edit;,,:, ' J

Vol. 9, No.5 B

Assis!ant [!dJtor' } ,

News of Computers and Data Processors:
ACROSS THE EDITOR'S DESK

CONTRIBUTING EDITORS'"

.

D: BOOTH

ANDREW

.

.

.inserted between pages 16 and 17

NBD CHAPIz:.1 "
JOHN

CA~R, III
. "
ALSTON S. HOUSEHOLDER:',
.. '
+"
,

W.

ADVISORY COMMlTI'EE
MORTON
,

M.

ASTRAHAN
,HOWARD

Rapid Indexing of Thousands of Chemical Articles

T.

ENGSTROM

W.

HAMMING

GEORGE

E. FORSYTHE

ALSTON

S. HOUSEHOLDER
HERBERT F. MITCHELL, JR.

SAMUEL

B.

RICHARD

WILLIAMS

ADVERTISING REPRESENTATIVES

Middle AtlantiC' States

L. KAYE'
535 Fifth Ave.
New York 17, N.Y.
MUrray Hill 2~4194

Washington

6; D.C.

ROBERT CA~EL

COlumbia 5-9727

A. S" BABCOCK
YUkon 2~3954
,W. F. GJU3EN
DUnkir~ 7·8135

San Francisco J'
605 Market'St:
tos Angeles .5 '
439' S. We.stern Ave.
,

"Elsewhere, .'

,

,

ARTICLES
Computer Programming for Command Control Systems, M. O. KAPPLER

8

Using Computer Services in Small Business, 1. J.
SELIGSOHN

14

READERS' AND EDITOR'S FORUM
Working Group For Better Education .

18

Thinking By Machines

19

Computer Bergerac, lYfILTON WEISS

20

Calendar of Coming Events

26

THE Pt;1BLlSHEll

'Berkeley Enterprises, Inc:
815 Washington St., Newtonville 60, Mass,
DI!cJltur 2~5453 or 2·3928

COMPUTERS and AUTOMATION is published monthly
at 815 Washington St., Newtonville 60, Mass., by
Berkeley Enterprises, Inc. Printed in U.S.A.
SUBSCRIPTION RATES: (United States) $7.50 for 1
year, $14.50 for 2 years; (Canada) $8.00 for 1 year, $15.50
for 2 years; (Foreign) $8.50 for 1 year, $16.50 for 2 years.
Address all Editorial and Subscription Mail to Berkeley
Enterprises, Inc., 815 Washington St., Newtonville 60,
Mass.
ENTERED AS SECOND CLASS MATIER at the Post
Office at Boston, Mass.
POSTMASTER: Please send all Forms 3579 to Berkeley
Enterprises, Inc., 815 Washington St., Newtonville 60,
Mass.
Copyright, 1960, by Berkeley Enterprises, Inc.
CHANGE OF ADDRESS: If your address changes, please
send us both your new address and your old address (as
it appears on the magazine address impri~t), and allow
three weeks for the change to be made.

4

1, 18

MILTON

1519 Connecticut Ave.

~

FRONT COVER

REFERENCE INFORMATION
Books and Other Publications, M. M. BERLIN

24

Survey of Recent Articles, M. M. BERLIN

27

New Patents, R. R. SKOLNICK

29

INDEX OF NOTICES
Advertising Index

30

Back Copies

see January, page 5

Computer Directory

see March, page 5

Manuscripts

see January, page 15

Reference and Survey Information
Who'~

Who Entry Form

.

see January, page 5
see March, page 24

COMPUTERS and AUTOMATION for May, 1960

omputer
Programmers:
Tired of standing on
the sidelines?
If you are content to work for instead of with
other staff members, System Development
Corporation is not for you. But, if you are ready
to come off the sidelines and get in the thick of
things, you should definitely consider SOCwhere programming is a primary function rather
than a service activity.

In addition to developing large computerized
control systems for SAGE, SAC, and other
important operations-SOC is engaged in a
number of long-range research projects. They
include: automatic coding and problem-oriented
languages; development of a language to
automate transition from one computer to
another; study of the organization of large
systems; investigation of computer design from
a standpoint of programability rather than
engineering; information retrieval and medical
data processing.
Positions now open at all levels
(at Santa Monica, California and Lodi,
New Jersey)
The extension of SOC's programming activities
into new areas has created openings for
Programmers at various levels of experience,
including senior status. Please send your inquiry
to Mr. E. A. Shaw, SOC, 2406 Colorado Avenue,
Santa Monica, California.
"Project CLIP- The Design of a Compiler and
Language for Information Processing;' a paper
by Harvey Bratman of SDC's Data Processing
Research staff, is available upon request. Send
request to Mr. Bratman at SOC.

SYSTEM DEVELOPMENT
CORPORATION
...
U1

COMPUTERS and AUTOMATION for May, 1960

Santa Monica, California • Lodi, New Jersey
5

ERINa THE FUTURE

ANNOUNCING

NEWEST ADDITION TO THE
HONEYWELL LINE OF TRANSISTORIZED
DATA PROCESSING SYSTEMS

Compare Honeywell 400 with any
other moderately priced system
•

Highest internal speeds (up to 10 times
faster)
• Fastest printing (with clean carbons)
plus 100% simultaneity with other operations
• Most reliable magnetic tapes (simultaneous read -write at 96,000 decimal
digits per second)
• Highest sorting speeds (two to three
times faster)
• Compatible with powerful Honeywell
800 (both program language and information transfer)
6

Just 14 months ago Honeywell excited the business
world with the introduction of the all- transistorized
Honeywell 800. Here at last was a system so compact,
powerful and efficient (and so low in cost) that it cast
an entirely new light on the economics of EDP.
Greater power was available to more users at lower
cost.
Today Honeywell announces a new low-cost system
that extends the benefits of magnetic tape data processing to more companies than ever before. This system is Honeywell 400, a low-cost complement to the
powerful Honeywell 800.

Solves the problem of growth

With the 400, Honeywell gives data processing committees another good answer to their problems. For here
is a system that is powerful enough to be economical, and,
at the same time, compact enough and priced low enough
to make it practical for many companies right now.
ThQ ample aapacity of the basic Honeywell 400 system
COMPUTERS and AUTOMATION for May, 1960

will accommodate reasonable growth. And, importantly,
when still greater processing power becomes necessary,
the graduation to a Honeywell 800 is a very logical and
economical step.

in touch with your nearest Honeywell Datamatic Sales
Office. Or, write to Minneapolis-Honeywell, Datamatic
Division, Wellesley Hills 81, Massachusetts.
Honeywell Datamatic Division Sales Offices in: ATLANTA

Compatibility is inherent

BOSTON. CINCINNATI. CHICAGO. CLEVELAND • DETROIT
HOUSTON • LOS ANGELES • NEW YORK • PITTSBURGH
PHILADELPHIA • SAN FRANCISCO • WASHINGTON, D. C.

Information can be readily interchanged between
Honeywell 400 and Honeywell 800 systems because they
use the same magnetic tapes (as well as the same highspeed printer and card reader). Most important, a programming language is provided that can be processed by
either system. When it becomes desirable to step up to a
still more powerful system, there is no need for costly
reprogramming. This compatibility also means that
Honeywell 400 programs can be checked out on Honeywell 800 systems, an advantage that could hasten the application of work to your Honeywell 400 by as much as
six months to a year.
For more facts about the new Honeywell 400, just get
COMPUTERS and AUTOMATION for May, 1960

In Canada: Honeywell Controls Limited, Toronto 17, On t.

Honeywell
g e~ Va:a 'P/I8etM~
7

COMPUTER PROGRAMMING FOR
COMMAND CONTROL SYSTEMS
M. O. Kappler
President, System Development Corp.
Sank! Monica, Calif., and Lodi, N.}.
(Based on a talk given to members of the U. S. Air Staff, Washington, D.C., October, 1959)

What is the role of computer programming in command control systems?
Let me first explain briefly what I mean by "command control systems." There are military organizations which have a job to do. This may be a job of
logistics, of combat, or both. As long as the operation
takes place within the sight and sound of the commander
or leader of such an operation, he can see what's going
on, and he can call over to the person he wants to
direct. But this is not the kind of situation that we're
confronted with in modern warfare; we're confronted
with a situation where everything is out of sight, a situation where a much larger and much less visible operation
takes place. The complexity of the operation demands
that information be reduced to symbols. These symbols
have to be "communicated in" to the commander or
command group; records have to be made; displays have
to be made; decisions have to be made; and commands
have to flow out. You notice I haven't said anything yet
about computers. In the concept I am setting forth, a
command control system exists whether you have computation, or data processing, or not. Let me say at this
point that computers or data processors are not necessarily ends in themselves; they are tools which aid us in
performing operations. When I refer to a computer, I
am referring to any kind of calculating machine, 'with
its associated display equipment and in-out equipment,
which utilizes a computer program.
Data Processing Assistance
Now, why do we even consider adding data processing assistance to a manual command control system?
There are several reasons. There is an overwhelming
volume of data that modern command control organiza-tions must deal with. The speed with which the data
must be reduced - made meaningful - is so great in
some instances that man cannot do the job manually.
Also, there is too much material to remember, even with
the aid of manual displays and recordings. In addition,
the decisions that have to be made, which are based on
this enormous amount of data, are too complex for the
unaided individual to make. Advances in technology
give us no alternative but to add data processing assistance or computer assistance to a manual command
control system. It is impossible for us to live in the
modern warfare environment without this kind of help.
Powers of Data Processing Equipment
The various kinds of data processing assistance presently available can greatly supplement the activities of
a control organization. Let's look for a moment at what
8

data processing equipment can do, just in general terms.
It can "communicate in," that is, it can receive messages
from components of the system, and it can translate
these into various forms for more useful application,
either for "communication out" or for storage inside.
It can delay them for further use until something else
is ready to utilize the information, and it can route
them around internally. This is the communication
function that it can perform "in." It can also serve as a
memory. It can keep things stored internally in an invisible form, or it can keep status boards that are visible
to the people who need the information. It can automatically update these status boards or the invisible
parts of the memory, and it can sort according to
various criteria - all that are longer than something,
shorter than something, faster than something, or of
higher priority than something. And it can correlate the
various things that are in the memory. It can also compute (I define computing here as the solution of mathematical equations). Furthermore, in some sense it can
even make decisions. It can make an optimization or a
sub-optimization, and it can make the kind of analytical
decisions which are difficult for people to make. A
moment ago I said that the kind of decisions that have
to be made in modern warfare are too complex for the
individual to make quickly and easily. The machine can
assist with this kind of decision. Also, the machine can
make various kinds of displays. These displays can be
made from the memory, either directly or from aggregates of information accumulated inside the memory, or
it can produce them from an analysis or computation
that it has made. I should mention also that the machine can make displays either continuously or on demand. Lastly, the machine can communicaate instructions based on the computations it has made.
Powers of Analog Computers and Digital Computers
The earliest data processing machines used by the
military, I believe, were analog machines. Analog machines are quite sophisticated from the standpoint of
mathematics. They can do calculus, for example. Analog machines in general, have the mathematics built in,
but digital machines, at least the ones we commonly
use, do not. We feel that the general purpose digital
machine is more advantageous than the analog machine
because we can make it a special purpose machine by
writing a computer program for it. Furthermore, we
can write more than one computer program. When we
write a program to solve a particular equation, the machine has that special purpose. Next week, when we do
another, the machine has a new purpose. In the case of
COMPUTERS and AUTOMATION for May, 1960

WORKING
PARTNERS
RCA 501 - SOUNDCRAFT INSTRUMENTATION TAPES

Big business depends more and
more upon electronic data processing. For
many corporations, the heart of their data reduction and storage operation will be the new RCA 501 Computer System.
The crucial testing period of this new computer called for
the most reliable of instrumentation tapes ... Soundcraft.
And, Soundcraft Tape proved to be the perfect working
partner-not only in the testing, but afterward. in continuous working use.
In short, experience has proven that Soundcraft works best
on leading computer systems, like the RCA 501. Let precision-made, trouble-free, error-free Soundcraft Instrumentation Tapes go to work for you. Complete literature on request.

REEVES

SOUNDCRAFT CORP.

COMPUTERS and AUTOMATION for May, 1960

Great Pasture Rd., Danbury, Conn.• Chicago: 28 E. Jackson Blvd.
Los Angeles: 342 N. LaBrea • Toronto: 700 Weston Rd.
9

the command control system the practice has been to
write a set - perhaps 25 or 30 computer programs for
a machine. Here the general purpose digital machine
becomes 25 or 30 different special purpose machines
solving different problems or carrying out different data
processing functions. Sometimes the machine runs
through a whole sequence of these in a few seconds.
However, there is no necessity for it to go through this
sequence in a fixed order. The order can'vary according
to the needs of the system. Also, in some systems it won't
be necessary for it to go through the cycles as quickly.
Depending on the demands of the system, it may be permissible for operations to take minutes rather than seconds.
Another aspect of the utilization of computers in command control systems, as distinguished from purely scientific computation, is the fact that the computer action is generally completed before the militaary action
and can influence the progress of the military action.
Some Command Control Systems
Lef us turn from these generalizations to look for a
moment at some specific, military functions that are
assisted by data processing. One of the command control
systems currently under development is the Strategic
Air Command Control System (465L). Earlier I mentioned the phrase "communicate in." One of the "communicate in" functions in the SAC Control System is to
receive data about logistics status in an emergency war
operation. This includes such things as aircraft availability, weapons availability, base capability, and weather.
These are all items which are received by the machine
and dealt with in some appropriate way. The SAGE
System (Semi-Automatic Ground Environment System
of Air Defens,e) differs slightly in that radar information flows automatically into the machine. This, by the
way, provides an example of my earlier reference to delay. Radar information has its own time pattern, which
corresponds to the rotation of the antenna. When this
timing is not in phase with the timing of one's need for
information, the flow of information can be 'delayed
momentarily until various activities can re-combine.
I said earlier that the machine can also translate. In
the SAGE System translation is a part of the data processing function. Radar information is taken from the
range-azimuth coordinates of the radar and is automatically converted to the rectangular coordinates of the sector by the machine. When the radar information is
ready to be used, it can be added to the track data, which
has been previously recorded by the machine, and associated with an already-established track. These are
typical examples of the kinds of data processing assists
that can be done in the "communication in" area.
The memory function now becomes obvious. In the
case of SACCS, for example, the, force status is in the
machine's memory unit at all times. It is even displayed
continuously on the display surfaces for ready reference
by the commander, who can request supplementary information as he needs it. There is a vast amount of information ready in the memory. There is also invisible
information available in the memory unit for the machine to call on for its own processing operations.
Intercepting
Thus far, I have been discussing computation as a
10

data processing function. Let me make, this more concrete by citing an example from the SAGE System, that
of intercept computation. You will recall that one reason for adding data processing assistance to a manual
system is that some operations may be difficult or impossible for a person to do manually. In intercept computation the action progresses too rapidly for manual
operation. Also, the extremely complicated mathematical computation involved in bringing the moving interceptor to meet another moving body can be done
much better by a machine.
Assigning Weapons
Another type of analysis that you can ask a machine
to make, to cite another example from SAGE, is in
weapon assignment. Consider the situation where the
approach of a number of bogey aircraft has been detected. There are several bases available; there are
several types of interceptors on each base; and each interceptor class has several types of weapons available.
Making a reasonable distribution of weapons and interceptors is clearly a problem that a man needs some help
with. Even with paper, a pencil, and plenty of time, he
can't do a very good job. The machine can measure the
distances, calculate the times of intercept, look at the
volume in which bogey aircraft are present and make a
weapon assignment that is much more sophisticated
than men can make.
\"Xlriting UDemosthenes"

Now a few words about the difficulty of programming. Suppose I want to make a display on a panel 3.nd
I call on a man to make the display. I tell him to write
"Demosthenes" on the panel and he does it. However,
if I want the computer to make the display, my instructions have to be considerably more detailed. In the
first place, I have to tell the computer that I want the
first letter 10 inches from the left hand side and 14 inches
from the top. Then I have to say what the first letter is.
Next I have to say that I want the second l~tter 14
inches from the top and 11 inches from the left hand
side. And I have to say what that letter is - a very
laborious process. This is one of the difficulties with
data processing assistan'ce in the present state of the art.
Methods of communicating with the machine are not
optimum from our viewpoint. I can't just tell it to put
"Demosthenes" on the panel. I have to write out a
very, very detailed description - called a computer
program - of what I want done.
Unable to Write a Program for Some Cases
Unfortunately our knowledge of all military operations is not ,as complete' as my knowledge of how to
write "Demosthenes" on the blackboard. Military operations are in~nitely more complicated, and a great deal
of effort is required to deal with the data processing
problems they present. We p.ave to formalize what we
want to say and then say it with great precision, especially when we want to get down into the computer
language. This is so true in many cases that we are
unable to ever get the program written in a fashion that
will enable the machine to assist us. When this happens,
our usual procedure is to let the machine operate up to
a certai~ point. Then a man reads the machine's display
COMPUTERS and AUTOMATION for May, 1960

material, takes some sophisticated action, reports this to
the machine, and the machine carries on. We use the
man as a mechanism because we don't know how to design a machine that's clever enough to do this particular
job. I should add here that this man is not the commander.
There may be another way to approach this problem,
however, and that is to learn to communicate with the
machine, to get into the machine in some fashion different from the detailed computer program coding fashion
we now use. We don't know how to do this, but I believe it is the most fruitful avenue to pursue to improve
our methods of utilizing data processing iin military
systems.
Communicating Out
I want to comment briefly on "communications out,"
which I referred to earlier. "Communications out" do
not necessarily have the same form. It isn't necessary
that SAGE outbound communication all have the same
language or that they all be in the same form. For example, SAGE forward-tell from the direction centers to
the control centers uses one message form; SAGE crosstell to the adjacent sectors is in two forms: track crosstell and fighter hand-over. SAGE also "communicates
out" time-division data-link information direct to airborne interceptors in still another form. We should not
fall into the trap of deciding that all the outbound communications of any particular system should have the
same form. Nor is it necessary that we have a common
computer language. These are limitations that we can
do without.
Let me return to computer programming for a minute. I want to emphasize the fact that neither the computer program nor the machine carry out a military
operation. In our context what really happens is that
the succession of computer programs in the machine
serves the command control organization as data processing assistance. By the way, there is no black magic
about writing a computer program. The specificity that
I referred to requires great, great volume, and the fact
that we interleave sequencing programs together makes
it complex and perhaps conceptually difficult, but it is
an orderly process. In my example of the display, I
think that if all of us knew the details of the mechanism of getting the letter down 14 and over 10, we could
sit down and write the computer program for that particular display. However, there is a difficult problem
associated with specifying precisely the military activity
for which you want data processing assistance. This is
especially true where intuition and judgment are Important.
Changing Programs at Later Dates
I don't think we should worry too much about flexibility in the program of data processing assistance in
military operations. In the concept that I have advanced, if you decide before you have completed the
design and implemented the system that the items that
you've selected for data processing assistance were
wrong or had the wrong priority assigned them, there
is nothing that prohibits modification or substitution.·
Furthermore, even after the system is operational, the
operational functions supported by data processing can
be revised by changing the computer program.
COMPUTERS and AUTOMATION for May, 1960

Scientific Observation of Operations
In order to design a command control system which
includes data processing assistance, scientific observation
of the operation is critical. Much less is known about
how to do this than is known about how to design a
machine. However, integrating data processing assistance is an orderly process. The procedure for designing
a command control system should start with the preparation of the system operation description (a description
of existing facilities, the integration of data processing assistance, and the data processing components).
The first step here is to observe both the current operation and the current organization. Since the system
probably won't be implemented for approximately two
years, the new hardware and the new operational concepts which will be in use by the time the system is
turned on should be considered.
In the case of SACCS, for example, the impact of
changes in missile technology must be anticipated. Because of changes in weapon technology and defense
configurations, recommendations will have to be made
that will change both operational procedures and the
organization itself. The manual organization should
also be reviewed and revised so as to take maximum
advantage of data processing support. In the case of
SACCS, it took about eight or nine months to get the
first version of the system operation description on
paper. Even then, the people who worked on it were
hesitant about publication, feeling that there were still
too many unanswered questions.
Iterations of Operating Descriptions
However, the best way to develop a better version of
the system operating description is to write it down,
circulate it among the people who are working on it,
and let them interact with each other about it. Next,
the hardware people should review it and give their
reaction as to whether it is feasible from the hardware
standpoint. Then it should be sent to the command
using the system to see whether it meets their needs.
There is not time for enough of these iterations in
SACCS before the date for an initial operating capability, but a minimum of three or four iterations is required
to do even a reasonable job. However, iterations can
continue after the system is in operation. Certain
things will be fairly fixed because of the nature of existing hardware, but the general purpose computer,
made special purpose by its program, enables us to revise the data processing support being given the command control organization. The iterative approach to
this problem is not nearly as precise as desired. Iterations like this only converge; they don't go directly to
the solution. Much more research and development on
the technology for implementing control systems is
needed. There is one important tool, though, that is
presently available: simulation. We anticipate the
characteristics of the operating situation for the command control organization and simulate them, not
elaborately, but with simple devices. This enables us to
actually bring people in and let them perform the operation. We can tell from their reactions whether or not
there is a reasonable system operation description and
whether the data processing assistance serves their needs.
We get a much better reaction from them than we could
11

by any discussion, because in simulation we reconstruct a
real situation. Our experience with simulation in air dedense shows that even with fairly crude simulation the
"crew members" get so completely absorbed that they feel
they are in the real situation. This method provides excellent design review.
Time and Cost
I don't want to minimize the amount of effort required to do iteration, simulation, etc. This is an activity that involves several hundred people for approximately two years. While these techniques do not produce a really complete job, they produce something that
is adequate. I think this is a very good investment. In
the case of SACCS, the total amount that the Air Force
expects to spend for the system may be from 200 million to 300 million dollars. The effort to do the system operation description will be approximately 10
million dollars, a small piece and certainly well worth
the effort. The computer program preparation will
probably cost about 10 million dollars.
Hardware
After the system operation description has been prepared, the earliest priority goes to the hardware designers who prepare the specifications for the equipment.
Since the hardware designers are needed for this purpose even before the first version is available, they must
interact with the engineers on a day-to-day basis in the
early phase of development. This interaction, which is

maintained through successive iterations, gives rise to
the methods and procedures that each individual crew
member will use, whether he is one of the mechanisms
in the system or whether he is in the command organization itself. It also yields the organizational design referred to earlier. It gives rise to plans and specifications
for both system training, that is, training the total
crews in conducting their activities, and individual
training. Lastly, it supplies the specifications for the
computer program that makes the general purpose machine into the special purpose machine required.
Making Computer Programs
A useful analogy to computer programm.ing is that of
a military aircraft and the operations orders that send
it out on a mission. In this analogy, the aircraft is the
machine and the computer program is the operations
order. If the commander is on the mission and he discovers that the operations order isn't satisfactory, he
can. modify it. Similarly, if the data processing assistance
isn't appropriate, it can be changed - both before and
after the operational date. There was a case of this kind
in SAGE. At the very last minute the need arose for
automatic target-to-battery evaluation for the NIKE battery. The data processing function was compressed a
little and A TABE became operational in the field in a
matter of six or eight months.
The computer programmer can't prepare the program
from written specifications. He has to become much more
knowledgeable and the correct practice, we believe, is for

the computer programmers to take their place in the team
writing the operation description system during the
early phases of development and proceed from there
after the first version.
People for Making System Operation Descriptions
The kind of people required to prepare the system
operation description are an unusual breed. They aren't
just engineers, but people who are oriented towards
and who are accustomed to looking at military operations. Many human factors specialists (psychologists),
for example, in their examination of operations for the
purpose of designing training programs, have gained
good insight into operations leading to a systems operation description. We have found that mixed teams of
engineers, physicists, human factors specialists and operations research people, who combine the knowledge of
the empirical and the less tangible, work out very well.
The training in statistical observation of human factors
specialists and operations researchers makes them especially valuable.
Exchange of Information Between Systems
I want to give some attention now to interface problems - the exchange of information between systems.
As we have seen in connection with SAGE "communication out," the technical problem of inter-communicating
between various systems is difficult but not insurmountable. We've made mistakes in this area in the past. For
example, when the FST-2 data converter at the longrange radar sites and the SAGE Direerion Center input

was first conneered, the FST-2 was reversed for entry
into the SAGE computer. Incidentally, these were not
programmed machines but logically wired machines
which had to be rewired. However, this is a much less
difficult problem than the question of operational requirements for communicating between systems. Let
us look for a moment at three systems: 438L, 425L and
4651. One system, 438L, has intelligence information
which it might just tell to 425L and 465L - and in
exactly the same form. The machine's conclusion to
use the same form would be wrong because SACCS has
quite a different need for intelligence information for
targeting purposes than NORAD does for defense purposes. The most difficult problem here is to examine
the operation, parcel out the operational functions to
one system or the other, and then write a very precise
description that will enable each operator to understand what comes in and what goes out. This is quite
different from the technical problem of common language. We should not be overly preoccupied with problems like "message form." These are very small details
compared with the operational problem of where the
information is. If we make a mistake in this area, however, our flexibility and our successive iterations will
provide relief. The iteration method applies not only
to the development of the single systems but applies
equally well to the way the systems will work together.
If we find that we've made an unsatisfactory division of
functions, we can adjust this to some extent after joint
operations begin.
The selection process of a contractor for one of

PHILCO 2000 DATA PROCESSING SYSTEM

World's First in All-transistor Logic and Circuitry

PH I Lea

I!I;I yamotlJ

tor Qualit;; Me It/ortd @vcr

PHILCO CORPORATION. GOVERNMENT & INDUSTRIAL GROUP
COMPUTER DIVISION • 3900 WELSH ROAD, WILLOW GROVE, PA.

these systems ought to place primary emphasis on the
capability of doing the operational job, not on a hardware proposal. It is a great temptation to view favorably a proposal in which there are tangible things. But it
is a mistake to select on the basis of hardware, since the
bidders who prepare proposals for us have spent only
two or three weeks, usually under pressure, and the
job of writing a system operation description is 100
times bigger than writing a proposal.
Research Needed in Data Processing
On the question of the research needed on better
ways of communicating with the machines, SDC is doing some of this work as a part of both its SACCS and
SAGE work and our other numbered L-systems work.
However, not nearly enough attention is being given to
this kind of activity. The problem of getting from a
word description into the machine needs much, much
more investigation. I hope that SDC and the Air Force
will jointly support more research here. The computer
manufacturers and the communications manufacturers
are, in some ways, outdistancing us. They can turn out,
in 'production, bigger and faster machines than we can
use. For our part, we don't have people to do system
operation description and write the consequent computer
programs in anywhere near the numbers necessary to
write an adequate system operation description for all

the systems of current interest, let alone for all the machines the manufacturers are capable of turning out.
There is some current research on computer programming, particularly in the language area, but we should
not be overly encouraged by this. There is a long process of operations observation and step-by-step organization of observations before we actually get to computer
programming. The research being done on computer
language is only a tenth or so of the total data processing assistance research. It is good work and it is helping tremendously to get computer programs written,
but it deals with only a part of the problem. In spite of
the modest research effort in computer programming,
our know-how has advanced a great deal. The number
of people involved in writing the SACCS computer program is only a tenth of that used in writing the original
SAGE program. Our learning curve in this area is good,
but we have a long way to go. We've made almost no
inroads into the problems involved in writing the system operation description. In addition to the need for
research in this area, there's room for quite a bit more
research in simulation techniques.
We can continue to devise good systems for the Air
Force to use, but the efficient implementation of these
systems depends on the successful prosecution of research in data processing which supports military operations.

Using Computer Services In
Small Business
I.

J.

Seligsohn

Computer Services Division
C-E-I-R, Inc.
Arlington, Va.
(Based on a report of the Small Business Administration, Washington 25. D.C.)

Only very large corporations can afford to purchase
a computer installation costing $2 million or more. And
even these corporations must keep such expensive equipment continually busy, usually 16 hours a day at least,
in order to payoff the initial heavy investment within a
reasonable time.
But what about the small businessman - the one with
say less than 250 employees? He needs a computer's
help scarcely once a week, perhaps only once a' month.
Can he take any advantage of the time-saving, labor-saving, giant electronic computers to strengthen his competitive position? For these machines have the lowest cost per
unit operation performed, once a certain minimum
number of operations is exceeded.
The answer is definitely YES - if the small businessman will make use of available computer services organizations.
Some computer service firms have complete staffs of
mathematicians, economists, and other scientists, in ad1'4

dition to their computer specialists. One (or a team) of
these people can come right to a small plant, analyze a
problem, and gather the data needed to solve it. This
does not mean, however, that a small manufacturer
with daily 'engineering problems can do without engineers of his own. Computer services experts should be
used as consultants on difficult problems, not as operating personnel. The small business owner who fails to
recognize this distinction is not being fair to himself
nor to any computer service with whom he may deal.
A further point is worth noting, toO : You can't just
walk into a computer services organization and get answers to all your problems right off the shelf. Many significant applications are already in use. There is no
doubt of that. But many others are more in the realm
of future possibilities; their greatest impact is still to be
felt. Nevertheless, the small manufacturer who fails to
look into computer services may be likened to the woodsman who felt he was too busy to sharpen his axe.
COMPUTERS and AUTOMATION for May, 1960

Computer Services
Typically, a computer services firm has one or more
large electronic general purpose digital computers on
its premises.
The general-purpose nature of the digital computers
implies that many different businesses can solve many
different kinds of problems on these machines. This is
exactly the case. Computer services organizations have a
variety of clients, many of them small, in nearly every
field of business and industry.
If you, as a small businessman, consult a computer
services firm, you will be offered two major types of
services to help solve your problem: programming
services, and computing machine services.
Programming Services
First, a computer analyst carefully studies your problem, and tells you whether a computer solution is
feasible and economical; in your situation, he helps in
determining whether the problem that brought you to
the computer services organization is complex or difficult enough to warrant a computer solution. When the
analyst has finished his work, he turns the problem over
to computer "programmers."
You pay for only the programming services for that
particular problem. You have no overhead costs of
maintaining a programming staff just to solve an occasional problem.
Computing Machine Services
When the problem is ready to be run on the computer,
both program and data are fed into the machine. The
computer then solves the problem at electronic speeds.
Answers can be put down on paper by a high-speed
printer.
You are charged only for the time it takes the computer to process your set of data. This charge usually is
based on an hourly rental of the computer. If you are to
be a large user of computer time, rates are often figured
on some other preiodic basis. Most small businesses, of
course, rent time on a com purer by the hour. They want
the benefits of fast, accurate computer solutions, without the large overhead of buying or renting their own
com outer full-time.
The computer services firm does, of course, have a
large overhead. But it is divided among many jobs and
clients. Often, computer services organizations plan to
keep their machines running two shifts a day, sometimes
around the clock. Thus, with fixed overhead spread
thinly over many clients, individual job costs can be
lowered.
What Types of Problems?
Large-scale, general purpose digital computers can
solve a wide range of business problems. The following
are representative:
Engineering Problems
Complex engineering problems often must be solved
before a given product can be manufactured. Typical
engineering problems that can best be solved by computers include stress analysis, heat and pressure calculations, vibration analysis, engine and lens design. Most
important of all, small manufacturers can now enter
new fields that were closed to them before the advent of
computers.
For example, one small manufacturer makes arresting
COMPUTERS and AUTOMATION for 'May, 1960

gears for jet planes. Recently he used a computer for
the first time to engineer his product. Then he announced enthusiastically, "From now on, in research
and product development - all the way from deep freezers
to atomic engines - the small company can hope for the
first time to make a definite contribution."
Lengthy Data-Processing Problems
The high speeds of electronic digital computers today
make possible the rapid processing of great amounts of
clerical data. Computers can drastically cut the time and
costs spent on functions such as payroll processing, billing, shop-order writing, and sales analysis. Certainly,
large corporations can and do effect greater savings by
applying computers to these problems. Why not small
business?
Indeed, some of them are now finding it increasingly
profitable to turn to computer services firms for dataprocessing help. They find that fewer cle~ks can handle
much more work in less time and at a lower ultimate
cost with the aid of computer services. This is especially
true in the case of a small firm expanding its business but
not yet able to afford more clerical personnel or a move
to larger quarters.
Computer services can also provide small firms quickly with up-to-date control information for use in inventory
control, cost accounting, and the like. The freshness of
this information is often vital to the success of a business.
Here's an example: One small manufacturer now receives a complete sales and inventory analysis once a
week - sometimes once a day when volume is high.
This enables him to balance his operations, preventing
over-production or under-production, while keeping inventories low but still without the risk of running short.
For highly seasonal products, computers can predict
within reasonably safe limits the requirements for preseason and working inventories of finished items and
sub-assemblies. This helps to level out production
curves, lower costs, and obtain more efficient operation.
Before computers were available, this manufacturer had
to make production decisions using information that
was months old. His plans were based on estimates and
samples rather than on accurate facts.
Other types of business, of cou,rse, need up-co-theminute information; for instance, wholesalers, manufacturers agents, construction firms, small banks, and
brokerage houses. One computer services firm specializes in rapid data-processing for banks that do not have
their own computers. At 6 P:M. each day, the computing firm receives from various banks the posting, documents, new account data, figures on installment payments
received, and so on. At 8 A.M. the next day, the banks
receive the day's complete transaction journal, and every
5 days they receive new borrowers' payment books,
delinquent accounts lists, and the like - all prepared by
the computer. The result? The banks are able to give
excellent service to their customers, with fewer personnel and lower costs, and with less delinquency. Customer accounts of manufacturers (merchants, agents,
and bro~ers, too) can be handled with equal facility by
c?mputers.
Production Problems
Small manufacturers have been making increasing
15

use of computer services to solve complex production
problems, such as:
(1) Machine shop scheduling -- finding the number of each item to be produced by each station in a
machine shop, to minimize delays and overtime, and to
shorten delivery time;
(2) Production scheduling - designing schedules to
meet expected sales, and varying these schedules in accordance with sales fluctuations;
(3) Blending or mixing problems - finding the best
combinations of ingredients or raw materials; and
(4) Reduction of trim losses - forming, shearing,
slitting, or punching rolls and sheets of material to produce various sizes, while at the same time minimizing
waste.
The typical small manufacturer seldom has the knowledge or staff to calculate mathematically the most precise
solution to such problems. Until recently, the best he
could do was make an educated guess among many alternatives. And the odds were always against his making
the most economical choice. But now, through the use
of computer services, the best solution can be found
swiftly and accurately. A high-speed computer can examine all the factors, explore all the possible choices, and
come up with the answer that will yield the most profit.
Transportation and Distribution Problems
Illustrative of these problems is a business which operates a fleet of trucks. The manager wanted to find out
the most economical routes for them to follow. The
factors that had to be considered included among others
mileages, traffic congestion, toll roads, and load capacities. In just a few hours, a computer analyzed 10
million possible route combinations and found the best
ones. Without a computer, it was estimated that it
would have taken 20 years to solve that problem. Obviously during that time the trucker would not have
been able to save several thousands of dollars per year
as he is now doing.
Similarly, computer services can help manufacturers
find out how to ship their products at lowest cost and
yet meet marketing requirements.
Management Analysis Problems
Computer services have also helped small businessmen
in the field of management analysis and prediction. They
have solved such problems as:
(1) Deciding where to locate a new plant or branch,
taking into consideration the expected customer demand,
transportation, distribution and the many other factors
that affect costs and profits;
(2) Deciding whether to make or buy componentswhich items or parts to manufacture, which to buy, and
when;
(3) Evaluating bids - deciding which offering is
most desirable when prices, terms and conditions vary
among the bidders; and
(4) Forecasting sales and demand - performing market lesearch by determining sales probabilities in different geographical areas for different products.
Many of the techniques used to solve thsee problems
have been developed by the relatively new field of operations research which applies the methods of science to
the operating problems of business. Some computer
services organizations have operations research experts
on their staffs. This combination often has a great ap16

peal for a business which must solve complex problems
without hiring a permanent staff of technical specialists.
What Does It Cost?
What will it cost you to take advantage of computer
services -- to have your problem programmed and
solved on a computer? At one end of the scale many
lar,ge computer manufacturers have formed computer
service subsidiaries. They sell computer time at around
$650 to $700 per hour. Programming time costs an
average of $15 per hour.
At the other end of the scale, small independent computer services firms sell computer time (on the same
types of equipment as the big companies use) at around
$350 an hour. Programming costs average $15 an hour,
but can go down to around $8 an hour for simpler problems. Of course, for recurring problems, where existing programs can be applied to new data, there are no
programming costs. A charge is made only for computer running time.
Certain background facts are worth considering in relation to these broad cost ranges. For one thing, large
computers are so expensive to buy and become obsolete
so quickly that over 90 percent of all computers are
rented from the manufacturers, rather than purchased.
The rental cost alone can be $500,000 to $800,000 or
more a year. One computer services firm with two machines pays $1,350,000 a year in rent. For another
thing, the installation and start-up costs for a single machine run around $800,000. Finally, the operating cost
for a single machine averages an additional $235,000 a
year.
The important question for small firm operators is:
"What do compqter services cost in terms of what we
get out of them?" Many companies have found that
the answers they got led to profits which more than offset
the expense. In such conditions every delay in getting
the needed data amounts to a charge against profits.
Some small business owners even consider the typical
rates to be a bargain. As one man put it: "For a few
thousand dollars in computer service costs, I'm saving
many man-months of work, and I've boosted my profits
by $10,000 per year.
For Further Information
Businessmen interested in learning more about computers and their applications to business problems may
wish to consult the following publications. In keeping
with the policy of the series, this list is brief and selective. However, no slight is intended towards authors
whose works are not included.
Electronic Data Processing for Business and Industry,
by R. G. Canning, John Wiley & Sons, Inc., New York,

N. Y. 1956. $7.
An Introduction To Automatic Computers: A Systems
Approach for Business, by Ned Chapin, D. Van Nosstrand Company, Inc., Princeton, New Jersey, 1957.,
$8.75.

Special Report - Computers: A Delayed Revolution,
Business Week Magazine, June 21, 1958. Reprints
available from Reprint Dept., Business Week, 330 West
42nd Street, New York 36, N.Y. 50c each.
To make contact with specific computer services organizations, a good starting point is the classified telephone directory of one of the larger cities; consult such
headings as "Computing" or "Data Handling Systems."
COMPUTEItS and AUTOMATION for May, 1960

NEWS

of

Computers

~~ACROSS

THE

COMPUTERS
Volume 9
Number 5 B

and

Data

EDITOR'S

Processors

DESK"

AND AUTOMATION

MAY 2, 1960

Established
September 1951

Published by Berkeley Enterprises, InC., 815 Washington st., Newtonville 60, Mass.

fSCORE:

15 OUT OF 25 SOLID-STATE EDP SYSTEMS
NOW IN USE, AND TWO NEW TYPES

RCA Electronic Data Processing Div., Radio
Corporation of America, Camden, N.J.
Of a total of twenty-five solid-state
computer systems now in use, RCA has supplied
fifteen of them -- all the RCA 501, or mediumsized electronic data processing system. The
company has orders for close to 50 more on
hand.
Two new systems were announced on April
13: the 301, smaller than the 501; and the
601, which is larger. They are already in
demand. There were 35 orders on hand for the
301, even before its formal announcement on
April 13. Considerable interest also has
been shown in the 601; the first two of them
already have been ordered by the New Jersey
Bell Telephone Company and Southern Bell
Telephone Company.
Data on the RCA 301
The RCA 301 is a completely-transistorized, general-purpose electronic data processing system for the medium size or smaller
business or government organization. It is
designed on the building block concept. Virtually any combination of punched card, paper
tape or magnetic tape input-output and disc
file or magnetic tape storage is possible to
meet varying user requirements.
In the RCA 301, flexibility is made possible through multiple, fast input-output media, rapid access core memory, a highly flexible and powerful instruction vocabulary, and
high speed arithmetic capabilities on such
problems as linear programming and matrix operations in connecti"on with scientific computation.
COMPUTERS and AUTOMATION for MaYt 1960

A disc file or "juke box" memory unit
holds 128 discs with magnetic recordings on
both sides for a capacity of 4-1/2 million
data characters. Up to five such disc files
may be incorporated in an RCA 301 system.
The discs may be rapidly interchanged or
erased for re-recording.
Up to 12 magnetic tape memory units can
be included in an RCA 301 system, each capable of storing nearly 4 million data charac-

-- Here are magnetic memory discs of the new
RCA 301 Electronic Data Proces,sing System, designed to provide full-scale data processing
for small firms. This is the first data processing system to use magnetic discs -- similar to 45 rpm records -- for its "memory storehouse". It also can use magnetic tape on reels.
The discs are said to provide new convenience
in the handling and storage of information.

1B

Data on the RCA 601
The RCA 601 is one of the world's faste~
electronic data processing systems. Fully
transistorized, it is suitable for both business and scientific applications and is capable of handling with ease the paperwork load
of many of the largest corporations.
The high speed memory is expandable up to
262,000 numbers, letters or symbols. As many
as 64 high speed tape stations can be linked
to the computer for the system's main memory
storage. Each unit is capable of holding 20
million data characters. The system has a
memory cycle of 1.5 millionths of a second,
with tape speeds up to 120,000 numbers, letters and other symbols per second. The co~
puter can multiply eleven-digit numbers in
10 millionths of a second. The system can
make 666,667 decisions per second.
A multi-programming technique has been
designed to permit the automatic running of
any number of independently written programs
simultaneously, limited only by the system's
total memory. In parallel processing or control operations, the RCA 601 not only will
work on a job priority basis but can handle
changeable priorities.
The RCA 601 uses a "uni ve,rsal computer
language", making it completely compatible
with other computers. Data can flow freely
between the RCA 601, the RCA 301 and the RCA
501, and all three can be readily integrated
into communications systems connecting widelyseparated company activities.
NEW MOTOR OIL DESIGNED BY COMPUTER
Esso Standard Division, Humble Oil and Refining Co., New York 19, N.Y.
A new motor oil built with the aid of an
electronic c~puter and atomic radiation was
announced April 1 by the company.
The formula for the oil, called New Uniflo, is complex, and an IBM 704 was needed to
help prepare the formula. Radioactive tracers were used to keep track: of the molecules
as they were built into the formula.
Comparative research with the new oil
and other very good lubricants showed that it
keeps car engines cleaner than any of the
others, Esso said. The research studies consumed two-and-a-half years of effort and thzee
million miles of testing in cars, including
road testing by a fleet of 150 taxicabs in
New York City. Stop-and-go travel like taxi
driving is extremely severe for engines because incompletely burned particles of gasoline collect to form sludge.
The new formula oil attacks the problem
of car engine lubrication and cleanliness in
a different way. It concentrates on breaking engine-formed sludge into minute partic~.

2B

far smaller than particles normally found in
an oil. These particles, less than a millionth
of an inch wide, Esso said, are kept floating
in suspension, away from engine parts they
would otherwise damage. When the oil is
changed, the sludge is drained off with the
oil, leaving the engine clean and assuring
longer life for engines.
At the heart of the new formula is a conr
plex polymer--or super molecule--on which most
of the computer and tracer activity was centered. Three ingredients have to react with
each other in a certain sequence to form the
polymer. The polymer is so intricate that
the computer was used to select the most
effective combination of the three ingredients.
To find out whether the computer was r~ht
the ingredients were tagged with Carbon 14--a
radioisotope--so they could be traced through
the reaction. This research confirmed the
704's predictions.
The three million miles of testing de~
onstrated the new oil's anti-sludge characteristics.
This development is believed to be the
most prominent role ever played by a computer
in developing any new lubricant.
HELICOPTER AIR-BORNE
ELECTRONIC AIR DEFENSE SYSTEM
Hughes Aircraft Co., Culver City, Calif.
Here is a versatile, electronic air defense system installed in five 3,500-pound
aluminum shelters transportable by helicopters.
The United States Marines has purchased it from
Hughes Aircraft Company, Fullerton, Cal., who
developed the "Airtrac" system, which gives air
defense capability to ground troops. Below,
one "helihut", transported as far as it can
go aboard truck is lifted skyward by a HR-2-S1
helicopter for movement to an isolated forward
area. The air defense system is operational
wherever these helihuts can be landed -- from
the bottom of a deep canyon to a mountain top.
They could go into action within minutes of
their set-down.

COMPUTERS and AUTOMATION for MaYt 1960

TEACHING DEVICE FOR REDUCING TRAINING TIME

SALE OF A DIVISION OF CLARY TO
REMINGTON RAND

General Atronics Corp., Bala Cynwyd, Pa.
A new automatic teaching device designed
to increase the effectiveness of the instructor in a wide variety of learning and training situations was announced by this company
at the end of March.
Called the Atronic Tutor, the new machine
is a simple, reliable, entirely mechanical
unit about the size of a small adding machine.
It presents step-by-step text material to the
trainee at a rate corresponding to his ability
to absorb it. The Tutor also tests the student on the subject matter, and measures his
progress.
This new technique, appears able to reduce the time required for industrial training courses by as much as 50 per cent ... or
teach the student up to twice as much in the
same period as is taught by normal teaching
methods.
The Atronic Learning System consists of
two parts:
(1)
The Tutor-Text, containing the subject
matter organized in a series of finely
graded steps or increments, with a
question or problem and a set of alternative answers on each page. The
text is written to build up a gradual
mastery of ~he subject as the student
moves from step to step.
(2) The Tutor, an inexpensive mechanical
device which turns the pages of the
Tutor-Text when the appropriate key
at its base is pressed. After the
student has read the information on
each page, he selects the answer from
the set of alternatives and presses
the corresponding key. If he is correct, the page turns; if he is wrong,
the Tutor waits until he chooses the
right answer. It takes less than
half a second for the page to turn.

COMPUTERS and AUTOMATION for May, 1960

Sale by the Clary Corporation of its
Adding Machine and Cash Register Division
assets to the Sperry Rand Corporation, Remington Rand Division, was announced in early
April by the two companies. The sale, which
is subject to final approval of the definitive contract by Clary stockholders and the
Sperry Rand Corporation Board of Directors,
will be for a cash amount of approximately
$8,050,000 and will become effective on
July 1.
The transaction will add a line of full
keyboard adding machines, and a line of manual and electric cash registers, to Remington
Rand lines. Included in the sale will be all
physical assets of Clary's manufacturing
plant in Searcy, Arkansas.
Clary personnel engaged in manufacturin~
sales, and service of adding machines and c~
registers, will be retained by Remington Rand.
Clary adding machines and registers will continue to be sold under the Clary name by the
present Clary dealers and other Clary sales
outlets.
Clary Corporation, at its headquarters
plant at San Gabriel, Calif.twill continue to
design and manufacture electronic computers,
electric print punches, and other data-handling equipment, and mechanisms for guidance
and propulsion systems used in missile
programs.
EDUCATING THE IBM 7070 COMPUTER
International Business Machines Corp., Data
Processing Div., White Plains, N.Y.
A group of fifty logicians, members of
IBM's Applied Programming Department, took
delivery in New York in late March of the
first IBM 7070 data processing system produced. Their assignment was to teach problem-solving short-cuts to the computer.
Operating around-the-clock continuously
for the next two months, the computer will
carry out more than ninety billion instructions, in order to determine and test programs for the new data processing system.
The high-speed computer and associated
equipment as installed at 11 Broadway is the
most powerful configuration of the 7070 data
processing system possible. This capacity
is necessary because of the initial applied
programming application. The system the IBM
programmers now command would sell for
$1,948,850. As comparison, a typical 7070
system sells for $900,500 and rents for
$20,350 a month.

3B

RPC-4000 ELECTRONIC COMPUTING SYSTEM
Royal McBee Data Processing Division, Port
Chester, N.Y.
A new, fully transistorized, generalpurpose, digital electronic computing system
was announced in March by Royal Precision Corporation, called the RPC-4000. It will be available at a monthly rental of $1,750, or a
total sales price of $87,500. This includes
the computer and the Tape Typewriter system.
First deliveries are scheduled for July.
The computing system is capable of operating on 9-digit numbers at rates up to 4,000
operations per second. The speed results from:
high operating rates; versatile command list
(42 in all) including commands which facilitate double-precision and floating-point operations; an index register that allows highspeed instruction modification; repeat execution feature; high-speed input-output equipment; and an 8-word accumulator for block
operation (eight sums may be accumlated simultaneously).
The magnetic drum memory section incorprates many new design features. The entire
unit is encased in a metal covering to protect it from dust particles or accidental
damage. The drum is tapered and floats on a
cushion of air. When the machine is idle,
the drum rests away from the heads. Air
pressure automatically raises the drum to
correct operating position when power is
turned on. The memory has a storage capacity
of 8,008 words (word length is 32 usable bi ts,
accommodating a 9-decimal digit number). A
variety of programs can be stored permanently
for instant access when needed. Average access time to the main storage is 8.5 milliseconds. Memory may be searched for full or
partial words (through special masking feature) at a sustained rate of nearly 200,000
words per minute.
Basic tape reader speed is 60 characters
per second, basic punch speed is 30 characters per second. A reversible photo-electric
reader, which reads punched paper tape at 500
characters per second, and a high-speed punch
(300 characters per second) are available as
optional accessories.
The computing system draws less than 10
amps., from a standard 110 volt line.
FAST NEW MINIATURE PRINTER
Potter Instrument Co., Plainview, L.I., N.Y.
Printing speeds of more than 30 characters per second on 5/16" tape are attained by
a small printer newly developed by this company. Only 9 1/4" long, wi th a self-contained
paper supply, the printer fits a standard 3"
instrument case.
4B

The high speed and small size of this instrument make it useful as a printer for computer output and for weapons system checkout
devices. It is also being used in burst communication to conserve the bandwidth of a radio transmission link. It relieves an aircraft pilot of the need to remember instructions and permits higher speed in control tower operation, thus reducing chances of pilot
error. It can provide a written record at
both ends -- in the plane and on the ground.
The new printer is now being tested by the
Federal Aviation Administration in connection
with the Air Traffic Control Program.
Other applications for the miniature
printer include: dispatching control for police cars and fire trucks, trains, busses,
taxi cabs and many other forms of transportation; communication systems for private planes;
office use to reduce the size of present office
teletype equipment, and by troops in the field,
using battery power and connected with a miniature keyboard.
Developed by Potter for Stromberg-Carlson,
this small high speed printer is designed to
sell for less than $500.00.

COMPUTERS FOR MEDICAL DIAGNOSIS
Packard Bell Electronics Corp., Los Angeles,
Calif.
Robert S. Bell, president of Packard Bell
Electronics Corp., told a luncheon meeting of
the Electric League of Los Angeles on March 21
that the day is coming when diseases will be
diagnosed electronically by advanced computers.
They will determine the disease and recommend
appropriate drugs and treatments according to
the patient's age, height, weight, and other
data and symptoms.

COMPUTERS and AUTOMATION for May, 1960

ANNOUNCE JOINT EFFORT IN COMPUTER CON1ROL
OVER POWER PLANTS
Republic Flow Meters Co., Chicago, and The
Thompson-Ramo-Wooldridge Products Company,
~ev~rly Hills, Calif.
These two companies have agreed to cooperate in designing and furnishing complete
computer-operated systems for electric power
industry. They will cooperate in engineering,
installing, and servicing complete valve, combustion-control, data-logging, alar~scanning,
and computer-c'ontrol systems to fulfill the
requirements of the power industry.
Republic has been a leader in pneumatic
combustion control systems for over a quarter
of a century; during the last 10 years Republic has designed and installed more than 125
electronic combustion control systems for
steam generating plants. Thompson-RamoWooldridge Products has been a pioneer and
leader in the development of industrial computer control systems for more than five
years, and has installed more than a dozen
of its RW-300 computer control systems for
on-line applications.
1HE HONEYWELL 400

ploit the economic advantages of automatic
parallel-processing, while the 400 is built
for situations where the volume of work is
insufficient to load an 800. However, it is
recognized that rapid sorting of data is always important and for this reason the Honeywell 400 provides the facility of simultaneously reading from one magnetic tape and
writing on another magnetic tape.
In addition to the memory, the arithmetic unit and the control unit, the central
processor contains the control circuits for
a printer, a card reader, a card punch and
up to six magnetic tape units.
Central Processor:
Speed
4500 to 6000 3-address operations per second
Memory
1024 words of core memory
(approx. 10,000 characters)
Checking Internal parity checking
Features Simultaneous read-write, special automatic editing provisions, high-speed sorting
ability
Options
Multiply-Divide, print-storage
for simultaneity with other
operations, additional
memory.

CO~UTER

Samuel D. Harper, Datamatic Division, Minneapolis-Honeywell, Wellesley, Mass.
The Honeywell 400, announced in early
April, although not as powerful as the Honeywell 800, is, nevertheless, a complete and
powerful system. First of all, it is a moderately priced all-transistorized data processing system, renting for under $lO-thousand a mopth; it is one Qf the most powerful
data processing systems available in its
price class. Because of its relatively modest price it will bring new-generation data
processing to thousands of companies that
could not afford the larger systems.
In standard form, it will look like this:
The Central Processor contains a memory
of 1,024 words, and operates with 48-bit
words. For comparative purposes this may
be thought of as providing storage for 10,000
characters.
Like the Honeywell 800, the 400 is programmed with 3-address instructions. This
is important not only on its own merits, but,
b~ing a specification of both the Honeywell
800 and 400, it facilitates the employment
of common programs.
The internal speed of the Honeywell 400
is approximately 6,000 of these 3-address
instructions per second. Compare this for a
moment with the speed of the Honeywell 800,
which is 30,000 3-address instructions per
second. The 800, however, is built to ex-

COMPUTERS and AUTOMATION for May, 1960

Magnetic Tapes:
96,000 decimal digits per
Speed
second
Features Identical to Honeywell 800
tape units;
Orthotronic Control (automatic error correction)
Printer:
Speed
900 lines per minute
Horizontal Span 160 columnar positions (any
120 may be active)
Features Clean carbons, ruggedness,
fully checked
Card Reader:
650 cards per minute
Speed
Features Fully checked
Other. Input-Output Devices:
Card
Punch
100 cards per minute or 250
cards per minute
Provision for accommodating other types
of input such as paper tape

5B

MULTIPLE USE FOR A UNIVAC 1105 COMPUTER
Remington Rand, Division of Sperry-Rand Corp.,
New York, N.Y.
A Univac 1105 computer system has begun
to work for education, industry, science,
government, and culture as a giant $3.6 million Computation Center was "unveiled" at the
end of March at the University of North Carolina, Chapel Hill, N.C.
Dr. Robert W. Burgess, director of the
U.S. Bureau of the Census, said that the Univac 1105 system will save time and money in
tabulating the 1960 census.
The census this year will be completed
in considerably less time than has been required to complete censuses in the past, he
said. "Our plans for the 1960 Censuses of
population and Housing contemplate faster
publication of results than has ever before
been achieved.
"We expect to have 100 per cent of the
data tabulated and published by early 1961~ .•
This will be 12 to 18 months earlier than
similar results for previous censuses were
made available.
Dr. Burgess said census officials "were
extremely pleased when we were able to work
out arrangements with Remington Rand, the
National Science Foundation and the University of North Carolina, for installations of
equipment exactly compatible with our two
Univac 1105's. Under the arrangement we are
guaranteed the ·part-time use of facilities
of this Computation Center during the period
when we most need help."
The Univac 1105 system, which is valued
at $2,400,000, was built by Remington Rand
and made available to the University of North
Carolina at a 50 per cent reduction in cost.
The other half of the cost was shared by the
National Science Foundation and the Bureau
of the Census, the latter agency purchasing
time on the Univac for the 1960 census. The
University of North Carolina constructed a
$1,200,000 building to house the Univac and
its accompanying Computation Center, which
is currently operating 20 hours a day on 15
projects.
COURSES IN COMPUTER SCIENCE AND ENGINEERING
University of Michigan, Engineering Summer
Conferences, Ann Arbor, Mich.
The following intensive courses for
practicing scientists and engineers will be
given June 13-24, at Ann Arbor:
Introduction to Standard Methods of Numerical Analysis -- Bernard A. Galler in
charge.

6B

Introduction to Digital Computer Engineering -- Harvey L. Garner and Norman R.
Scott in charge.
Theory of Computing Machine Design -Harvey L. Garner and Norman R. Scott in
charge.
Advanced Numerical Analysis -- R. C. F.
Bartels and T. B. Curtz in charge.
Programming Concepts, Automata, and Adaptive Systems -- J. H. Holland in charge.
Applications for enrollment and requests
for information should be directed to R. E.
Carroll, Engineering Summer Conferences, 126
West Engineering Building, Ann Arbor, Mich.
WESTERN JOINT COMPUTER CONFERENCE AND EXHIBIT,
The sponsoring societies of the Western
Joint Computer Conference are: Institute of
Radio Engineers; American Institute of Electrical Engineers; and the Association for
Computing Machinery. The conference will be
held May 3, 4, 5, at the Jack Tar Hotel, San
Francisco, Calif.
"The Challenge of the Next Decade" will
be the general theme at this year~conference
and the topics of the sessions and papers are
as follows:
Tuesday, May 3
Session I-A
COMPUTER ORGANIZATION TRENDS -- Panelists:
Gene ~mdahl, Aeronutronic Corp.; Morton
M. Astrahan, IBM Corp.; J. Wesley Leas,
RCA.
The Historical Development and Predicted
State-of-the-Art of the General Purpose
Digital Computer -- C. P. Bourne and
D. Ford, Stanford Research Institute
The Harvest System -- P. S. Herwitz and J.
H. Pomerene, IBM Corp.
Horizons in Computer Systems Design -- Walter
F. Bauer, Ramo-Wooldridge Corp.
Organization of Computer Systems--The Fixed
Plus Variable Structure Computer -Gerald Estrin, University of California,
Los Angeles
Session II-A
DATA RETRIEVAL -- Panelists: John Postley,
RAND Corp.; Mrs. P. B. Bremer, FMA, Inc.;
M. E. Maron, RAND Corp.
A MUlti-Level File Structure for Information
Processing -- Louis Miller, Jack Minker,
W. G. Reed, and W. E. Shindle, RCA.
Symbolic Logic in Language Engineering -H. M. Semarne, Douglas Aircraft Co., Inc.
The Fact Compiler--A System for the Extraction,
Storage, and Retrieval of Information -Charles Kellogg, Ramo-Wooldridge Corp.

COMPUTERS and AUTOMATION for May, 1960

Session II-B
COMPONENTS AND TECHNIQUES -- Panelists: D. A.
Meier, National Cash Register; Norman L.
Kreuder, Electrodata
A Word-Oriented Transistor-Driven Non-Destructive Read-Out Memory -- D. G. Fischer and
T. C. Penn, Texas Instruments, Inc.
Unifluxor: A Permanent Memory Element -- A.
M. Renard, Aeronutronic; W. J. Neumann,
Remington Rand Univac
Characteristics of a Multiple Magnetic Plane
Thin Film Memory Device -- K. D. Broadbent, S. Shohara, and G. Wolfe, Jr.,
Hughes Aircraft Co.
Session II-C
ANALOG EQUIPMENT -- Panelists: Vernon L.
Larrowe, University of Michigan; Mark
E. Connelly, Massachusetts Institute of
Technology
Analog Time Delay System -- Charles D. Hofmann
and Harold L. Pike, Convair Astronautics
DAFT: A Digital/Analog Function Table -Robert M. Beck and Jack M. Mitchell,
Packard Bell Computer Corp.
Mathematical Applications of the Dynamic Storage Analog Computer-- J. M. Andrews,
Computer Systems, Inc.
Wednesday, May 4
Session III-A
LEARNING AND PROBLEM SOLVING MACHINES -Panelists: Oliver Selfridge, Lincoln
Laboratory, MIT; J. C. R. Licklider,
Belt, Beranek, and Newman, Inc.; H. J.
Bremermann, University of California,
Berkeley
Recognition of Sloppy, Hand-Printed Characters -- Worthie Doyle, Lincoln Laboratory, MIT
Empirical Exploration of the Geometry Theorem
Machine -- H. Gelernter, J. R. Hansen,
and D. W. Loveland, IBM Corp.
A Suggested Model for Information Representation in a Computer th~t Perceives,
Learns. and Reasons -- Peter H. Greene,
The University of Chicago
Session III-C
ANALOG TECHNIQUES -- Panelists: G. A. Korn,
University of Arizona; Walter Brunner,
Electronic Associates, Inc.
Analog Computer Techniques for Plotting Bode
and Nyquist Diagrams -- G. A. Bekey and
L. W. Neustadt, Space Technology Laboratories, Inc.
On the Reduction of Error in Certain Analog
Computer Calculations by the Use of
Constant Equations -- R. M. Turner,
Lockheed Aircraft Corp.
The Use of Parameter Influence Coefficients
in Computer Analysis of Dynamic Systems
-- Hans F. Meissinger, Hughes Aircraft
Co.

COMPUTERS and AUTOMATION for May, 1960

Session IV-A
TRENDS IN COMPUTER APPLICATIONS -- Panelists:
Clair E. Miller, Electronic Computing
Center, San Francisco; A. R. Zipf, Bank
of America; George W. Evans, Stanford
Research Institute
Data Processing--What Next? -- John M. Salzer,
Ramo-Wooldridge Corp.
The Outlook for Machine Translation -- Franz
L. Alt, National Bureau of Standards
Computers for Artillery -- Lt. Colonel Louis
R. van de Velde, U. S. Army Artillery
and Missile School, Fort Sill
Session IV-B
LOGICAL DESIGN -- Panelists: A. Jennings,
California Computer Products; D. Aufenkamp, General Electric Co.; B. Elspas,
Stanford Research Institute
Communications Within a Polymorphic Intellectronic System -- George P. West and
Ralph J. Koerner, Ramo-Wooldridge
Encoding of Incompletely Specified Boolean
Matrices -- T. A. Dolotta and E. J.
McCluskey, Jr., Princeton University
A Built-in Table Lookup Arithmetic Unit -R. C. Jackson, W. H. Rhodes, Jr., W. D.
Winger, and J. G. Brenza, IBM Corp.
Session IV-C
HOW WELL ARE WE PROCESSING OUR OWN INFORMATION? -- Panelists: Sandy Lanzarotta,
Editor, "Datamation"; Jackson W. Granholm, Edi tor, "Computing News"; Fred .
Gruenberger, News Edi tor, '·'Communications of the ACM"
Thursday, May 5
Session V-A
DESIGN, PROGRAMMING, AND SOCIOLOGICAL IMPLICATIONS OF MICROELECTRONICS -On Microelectronic Components, Interconnections, and System Fabrication -- Kenneth
R. Shoulders, Stanford Research Institute
On Iterative Circuit Computers Constructed of
Microelectronic Components and Systems
-- John H. Holland, University of Michigan
On Programming a Highly Parallel Machine to
be an Intelligent Technician -- Allen
Newell, The RAND Corp.
On a Potential Customer for an Intelligent
Technician -- C. West Churchman, University of California, Berkeley
Session V-C
ANALOG APPLICATIONS -- Panelists: John H.
McLeod, Convair-Astronautics; Donald
F. Zawada, Ford Motor Company
Real-Time Automobile Ride Simulation -Robert H. Kohr, General Motors Corp.
Using an Analog Computer for Both Systems
Analysis and Operator Training on the
Enrico Fermi Nuclear Power Plant --

7B

Samuel N. Irwin, and Robert Kley, Holley
Carburetor Company
Anatran--First Step in Breeding the Diginalog
-- Lee Ohlinger, Northrop Corp., Norair
Division
Session VI-A
PROGRAMMING SYSTEMS -- Panelists: Robert L.
Patrick, Manhattan Beach, Calif.; Ascher
Opler, Computer Usage, Inc.; Richard
Ridgway, IBM Corp.
Man-to-Machine Communication and Automatic
Code Translation -- A. W. Holt, on
leave from Remington Rand Univac to the
Moore School of Electrical Engineering,
and W. J. Turanski (deceased)
The Computer Operation Language -- G. F.
Ryckman, General Motors Corp.
A New Approach to the Programming Problem -William Orchard-Hays, Corp. for Economic
and Industrial Research, Inc.
Session VI-B
INPUT-OUTPUT AND COMMUNICATIONS -- Panelists:
J. Svigals, IBM Corp.; G. Warfel, Bank
of America
A Line-Drawing Pattern Recognizer -- Leon D.
Harmon, Bell Telephone Laboratories
Automatic Store and Forward Message Switching System -- T. L. Genetta, H. P.
Guerber, and A. S. Rettig, RCA
The Videograph Label Printing System Developed for Time· Inc. -- B. H. Klyce, Time
Inc., and J. J. Stone, A. B. Dick Co.
EXHIBITS: Exhibits will be located adjacent
to the technical sessions in the Jack
Tar Hotel. These will be open to the
public as follows:
May 3
12 noon to 6:00 p.m.
May 4
12 noon to 9:00 p.m.
May 5
9:00 a.m. to 6:00 p.m.

SYSTEb~

June 15
A Simple Queue Simulator (I). Construction
of a computer program to simulate a
simple queuing system. Sidney Saltzman.
A Simple Queue Simulator (II). Simulation
runs will be made at the Cornell Computing Genter.
lbe Cornell Simulator. A program to simulate
a production shop. W. L. Maxwell, Cornell University.
June 16
Problems in the Design of a Simulation Experiment. Elements of experimental design
important in simulation. R. E. Bechhofer,
Cornell University.
Operating Problems in a Simulation Experiment.
Problems of multi-variable exploration,
equilibrium, starting conditions and
sampling. R. W. Conway.
Production Control Simulation (I). Experimental comparison of production control
decision rules suggested by seminar participants using the Cornell Simulator.
W. L. Maxwell.
Manufacturing Systems Simulation in the General Electric Company. W. E. Barnes,
Production Control Service, General
Electric Co.
June 17
Production Control Simulation (II). Analysis
and evaluation of experimental results.
W. L. Maxwell.
Survey and Summary of Applications of Simulation. General discussion and conclusions. R. W. Conway.
Informal Discussion. An optional period for
individual consultation and informal
group discussion.

SIMULATION USING DIGITAL COMPUTERS
SEMINAR
AUTOrMTIC DEGAUSSER OF MAGNETIC TAPE ON REELS

Cornell University, Ithaca, N.Y.
Consolidated Electrodynamics, Pasadena, Calif.
An industrial engineering seminar on
"Systems Simulation Using Digital Computers"
will take place at Cornell, June 14-17. The
program follows:
June 14
Introduction. The experimental investigation
of complex systems. R. W. Conway, Cornell Uni versi ty.
The Modern Digital Computer. Equipment requirements for digital simulation.
R. W. Conway.
Logical Representation of an Operating System.
Translation to computer language. Sidney
Saltzman, Cornell University.
8B

An automatic degausser for reeled tapes
has been added to this company's line of magnetic tape equipment. A reel of instrumentation tape recorded to saturation is erased by
the degausser to at least 50 db below normal
record level.
The new Type 5-055A Automatic Tape Degausser can be used with tapes from one-quarter to two inches in width, reels from seven
to 14 inches in diameter, and reel hubs of
all dimensions.
Uniform degaussing is achieved with automatic time cycling of reels in the magnetic
field. After a reel of tape is placed on the
turntable spindle and the OPERATE button
COMPUTERS and AUTOMATION for May, 1960

pressed, capacitor-type motors slowly move the
reel into the erasing field, rotate it several
times, and slowly withdraw it. When the reel
returns to its original position the degausser
shuts off. The degaussing cycle is completed
in approximately 60 seconds. Heavy shielding
prevents erasing field radiation from reaching nearby tape system components.
TWO RW-300 COMPUTERS WILL BE INSTALLED IN
SECOND FRENCH NUCLEAR POWER PLANT
Thompson-Ramo-Wooldridge Products Company,
Beverly Hills, Calif.
The second commercial nuclear power plant
in France will utilize two RW-300 Digital Control Computers to detect and locate burst fuel
slugs.
The pIa nt is being constructed near
Chinon by Electricite de France, electric po~
er utility owned by the French Government.
The nuclear reactor will be started up at the
beginning of 1961 with the two RW-300 computers connected to the system for monitoring
the radioactivity of the cooling gas.
Two RW-300 computers are now being installed in the first French commercial nuclear
power plant, also at Chinon, France.
The second plant will have a rated output
of 800 megawatts of heat and 250 megawatts of
electricity. It will be among the largest nuclear power plants in the world. The reactor
will require 250 tons of natural uranium as
fuel and 2,000 tons of carbon (graphite) as
moderator. Carbon dioxide at a pressure of
360 pounds per square inch will transfer heat
from the reactor to a heat exchanger, where
steam will be formed to drive turbine-generators.
Intertechnique, a prominent French electronics and instrumentation company, has the
prime contract for the reactor monitoring
system which will incorporate the two RW-300
computers. Intertechnique has primary responsibility for engineering the instrumentation
and data-handling portion of the burst-slug
detection system into the nuclear reactor
plant.

footage. Possibility of error is reduced by
as much as 75 per cent.
The chart, comprising 3,200 calculations,
is placed on top of a scaled map of an area to
be measured. Origin of the chart coincides
with the lower left corner of the rectangular
area to be measured. Exact square footage may
then be read in the diagonally opposite corner.
Calculations for the chart were worked
out on a computer in 13 minutes. Although the
chart was designed specifically for determining square footage on a one-quarter-scale layout, it may of course be prepared in any unit
of measurement, such as miles, yards, etc.
Such charts may find wide use on scaled-maps
used by architects, engineers, the military,
and others.

COMPUTER-DEVELOPED AREA CHART
General Electric Co., Phoenix, Arizona
By using a computer, two plant-layout
engineers at the General Electric Company's
Computer Department have developed a transparent, calibrated area-chart for determining
at a glance the square footage of any rectangular mapped surface. Time studies show use
of the new chart saves up to 40 per cent of
the time formerly required to determine square

COMPUTERS and AUTOMATION for May, 1960

9B

THE FERREED: MICROSECOND MEMORY DRIVES
METALLIC CONTACTS
Bell Telephone Laboratories, Whippany, N.J.
A team of engineers here has invented a very
fast electromechanical switch called the "ferreed".
It utilizes a new technique for actuation which is
compatible in speed with electronic circuits.
The ferreed may find use as an interconnecting
element in telephone switching networks, where it
could be controlled a thousand times more rapidly
than switches presently employed.
Inventors of the new switch, A. Feiner, C. A.
Lovell, T. N. Lowry, and P. G. Ridinger, are members of the technical staff of Bell Telephone Laboratories. They anticipate applications for the
ferreed in various types of systems where advantage
may be taken of metallic contacts controllable by
current pulses of a few microseconds duration.
Magnetic materials that can be quickly switched
between two alternate states have been widely used
as "memories" for storing information in digital
computers. The useful output from these elements
has generally been limited, however, to electrical
signals of a transitory character.
In conventional electromagnetic relays, on the
other hand, continuous electrical currents ca~be
used to open and close metallic contacts for extended periods of time, but at speeds limited by the mechanical motion of the moving parts.
The ferreed combines the rapid switching of bistable magnetic material with metallic contacts for
output indications that persist as long as desired
without further application of power. In several
mOdels of the device, a cobalt ferrite has been
used as the magnetic material and a glass-sealed
magnetic reed switch for the output contacts -hence the name, "ferreed".

In operation, the magnetic material is switched
by a magnetomotive force applied, typically, as a
five-ampere current pulse in a thirty-turn winding.
Control pulses as short as five microseconds will
switch the magnetic material, resulting in the passage of magnetic flux through the movable members
of the reed switch. Actual closure of the contacts
is delayed by inertia of the reeds for several hundred microseconds.
Release of the contacts is brought about by
cancellation of the magnetic flux through the reeds,
as the result of another five-microsecond switching
operation. Opening of the contacts requires less
time than the closing operation.
Several new magnetic materials have been synthesized for use in developmental models of the ferreed. Among these are ferrites exhibiting characteristics midway between permanent magnets and computer
memory materials, and ferrite suspensions in plastic
"tailor-made" for specific magnetic, electrical, and
mechanical properties. The reed switch is similar
to a type used in standard electromagnetic relays
for the past decade.
In a telephone switching network, subscribers
are interconnected by the closure, in specified
patterns, of switches called "crosspoints". A typical central telephone office contains many thousands
of these switches; of these, a few dozen are operated
for each call. Those crosspoints required for a
particular connection must be selected by the coincidence of pairs of control pulses. The ferreed's
magnetic structure responds to a pair of pulses by
closing the output contacts. They are opened by a
subsequent single pulse.
Besides coincident pulse operation and rapid
actuation, the magnetic structure permits the desired switching by current pulses of widely variable character.
The result, when ferreeds are used in switching networks, is reliable operation from simple
control circuits.

The ferreed, developed at Bell Telephone Laboratories, is a very fast electromechanical switch with metallic contacts, requiring no holding power.
In the two-branch ferreed pictured, two ferrite bars, two glass-sealed magnetic reed switches, and two plastic
end-pieces comprise the magnetic circuit.
Non-conducting electrically, the plastic end-pieces magnetically couple the ferrite bars to the magnetic reed
switches, at the same time providing a good structural support for the assembly.
The passage of magnetic flux through the movable metallic members of the reed switch closes a circuit. Control
pulses of five microseconds through the windings on the ferrite will switch the magnetic material.
lOB

COMPUTERS and AUTOMATION for May, 1960

RESEARCH ACTIVITY AT PHILCO
-- SOME NOTES
Philco Research Division, Philco Corp., Philadelphia 34, Pa.
On January I, 1960, Philco's Research group became a separate division, with its own administrative organization, preparatory to a move in early
1961 to a new Research Center under construction in
Whitpain Township, near Norristown, Pa.
According to Philco President James M. Skinner,
Jr., "We have spent twice as much on research in
1959 as was spent in 1949; we predict that research
expenditures will double again in the next five-year
period."
Areas of Research Activity
Basic Science and Technology:
Semiconductor metallurgy:
Single crystals
Polycrystallin~ materials
Magnetic and dielectric materials
Physics and chemistry of transistors
New physical and chemical sources of electric power
Thermoelectric cooling
Electron optics
Information theory applied to communications and radar
Devices and Components:
Transistors
Microelectronics and micromodules
Transistorized, computer components
Infrared devices
Single-crystal and film detectors, imaging
systems, cooling devices
Microwave components
Combination of transistors and harmonic
generators for higher frequency performance
Parametric-diode amplifiers
Semiconductor oscillators and amplifiers for
X-band and above
Thermoelectric power generators and fuel cells
for conversion of heat energy to electricity
Electrovisual devices -- data displays and
techniques for the manipulation of visual
data
Circuits and Eguipments:
Arithmetic and memory assemblies for computers
Digital circuits and equipments
Analog circuits and equipments
Acoustic subsystems
Digital computer circuits, continued work
toward higher speed
Conversion -- analog-to-digital, digital-todigital, digital-to-analog
Digital communication
Microwave circuits -- enhancement of signaldetection; solid-state devices for oscillation and amplification.
Systems Research:
Missiles and space systems
Surveillance -- radar and infrared
Communications -- ground and space
Anti-submarine warfare
COMPUTER FOR BRITISH GOVERNMENT PENSION OPERATIONS
F. C. Livingstone, London, England

scheme. The Ministry of Pensions has placed an order for an Emidec 2400, one of the largest and most
advanced computers in Europe.
This machine will check the records of every
insured person in Britain - some 25 million - every
day. The job will take about four hours.
Thirty million forms a year will be processed
when the scheme is running full blast. The relevant
information will be fed into the computer's memory,
the main part of which will consist of 22 large magnetic tape recorders. A new type of printing machine, which can print 50 lines a second, will put
the results of the machine's calculations down on
paper.
This is the second Emidec 2400 to be sold. The
first was bought by the Royal Army Ordnance Corps,
mainly to keep track of spare parts used in Army
vehicles.
An Emidec 3400 is being planned, a still larger
and more powerful machine than the 2400. It will be
on a similar scale to the Ferranti Atlas machine being developed in association with Manchester University and the IBM "Stretch".
The British Government's National Research Development Corporation has been supporting the development of the Emidec 2400 and 3400 and is also supporting the Atlas.
DEVELOPING AND TESTING BOOK-MARKING EQUIPMENT
Council on Library Resources, Inc., Washington, D.C.
Two grants, totaling $22,600, have been made by
the Council on Library Resources, Inc., to the American Library Association, 50 East Huron St., Chicago.
The grants are for research and development to
be undertaken by the organization's Library Technology Project.
A $20,000 grant will be used to develop a mechanical book-marking device to replace the present
hand methods. A smaller grant of $2,600 covers
several small testing programs which will be conducted by the Chicago Paper Testing Laboratory, Inc.
Battelle Memorial Institute of Columbus, Ohio,
will conduct research and development on the bookmarking machine. The millions of volumes added to
library collections each year must be individually
labeled on the spine with a number. Development of
a book-marking device would meckanize a process
which has used slow, unsatisfactory hand methods
for many years. A conservative estimate indicates
that with a satisfactory machine books could be
marked twice as rapidly as by hand and with increased legibility and complete uniformity.
The project hopes to develop a device similar
in size and ease of operation to a small adding machine. Battelle will conduct the project in two
phases, the first to demonstrate the feasibility of
the system, and the second to construct a complete
prototype. The present grant covers the first phase
only, which will take about six months.
If the first phase is successful, the construction of a prototype is expected to take about four
months. The completed prototype will be library
tested, and cost comparisons will be made between
present methods of marking and the machine method.
The Council on Library Resources, Inc., is a
non-profit agency established in 1956 with the aid
of a grant from the Ford Foundation to assist in
solving the problems of libraries.

An enormous electronic computer is to be installed at Newcastle-on-Tyne, England, to handle,
the British Government's new graduated pensions
COMPUTERS and AUTOMATION for May, 1960

llB

MAGNETIC CORE MEMORY TO WITHSTAND SHOCKS
OF 10 G AND VIBRATION OF 2000 CYCLES PER SECOND
Telemeter Magnetics, Los Angeles, Calif.
A new magnetic core memory stack for withstanding great vibration and shock has been developed by
this company. Made up of an assembly of core arrays,
the new memory stack is designed to withstand the
rigors of military applications where vibration and
shock would destroy ordinary units.
The new units, which are called MIL-STAK, are
unaffected by acceleration to 10 g with vibration
to 2000 cps.
To achieve the rugged construction, a molded
plastic frame has been designed on which the cores
are wired to form an array. Then the wired arrays
are interconnected by a locked link method and then
the entire stack is encapsulated in polyurethane.
With the interest in automatic digital control
of missiles and other space vehicles, a wide variety
of applications is anticipated.
CODE AND SWITCHING DRUMS USING EMBEDDED CIRCUITS
Beck's, Inc., St. Paul, Minn.
The accompanying figure shows some of the drums
carrying codes and switching circuits made by this
company. The shaft can be metallic or nonmetallic,
of any length. The conducting area can be copper,
brass, silver, aluminum, etc.; and can be plated.
The insulation can be epoxy or other resins with
or without fillers, and is flush with the conduct.fng areas. The inside of the drum can be hollow
or filled. The diameter can be one quarter inch to
four inches or larger. The code pattern can be any
that is desired. The code drums are used in telemetering, automation, control devices, etc.

HIGH-SPEED COMPUTATION AND INSTRUMENTATION
FOR PROCESS CONTROL
Instrument Society of America, 313 Sixth Ave., Pittsburgh 22, Pa.
Under the theme "High Speed Computing and Instrumentation for Process Control" a number of papers
are scheduled for the technical sessions of the Instrument-Automation Conference & Exhibit of the Instrument Society of America, to be held in the Civic
Auditorium, San Francisco, May 9 through 12, 1960.
Approximately 60 papers are scheduled for some
25 technical sessions of the Conference. The session dealing particularly with computing, is:
DATA HANDLING & COMPUTATION
(Wednesday, May 11 - 9:30 a.m. to Noon)
"Applications of the Quadratron to Computation
and Data Reduction", Dr. Ladis A. Kovach
and William Comley, Douglas Aircraft, Inc.,
El Segundo, Calif.
"The Generation of Information For Decisi.ons
Of Specified Risks", J. E. Witherspoon,
Rocketdyne, Canoga Park, Calif.
"Real-Time Aspects of On-line Computer Control", Arthur J. Donovan, Librascope
Div'n., General Precision, Burbank, Calif.
"Real-Time Digital Computer Control System For
Industrial Applications", Geert H. Bouman,
Minneapolis-Honeywell Regulator Co., Phildelphia, Pa., and Dr. Joseph J. Eachus,
DATAmatic Division, Minneapolis-Honeywell
Regulator Co., Newton Highlands, Mass.
PRODUCTION IMPROVEMENTS FOR THE NUVISTOR TUBE
Radio Corp. of America, Camden, N.J.
Several important new developments for the economical production of RCA's thimble-size nuvistor
electron tube have taken place:
(1) a brazing operation for the nuvistor's
assembly in which eleven wires are
sealed vacuum-tight through a ceramic
wafer while fourteen other joints are
made;
(2) because of the open-ended cantilever
construction within the nuvistor, complete mechanized assembly of the parts
into jigs for accurate positioning of
electrodes;
(3) means to control the processing of
vitreous ceramic bodies so that finished parts can be made to very small
tolerances without expensive machining
operations;
(4) a simple, reliable method for metallizing the ceramic.

12B

COMPUTERS and AUTOMATION for May, 1960

... General Electric's New $14,000,000 Space Research Center,
to be built. near Valley Forge Park 17 miles from Philadelphia
The Missile and Space Vehicle Department of General Electric-a recognized
leader in the development of instrumented re-entry vehicles -is now pursuing a number of even more advanced space programs. Basic to progress
in these programs is the solution of a diversity of interesting mathematical
problems. These include trajectory and navigation studies and analysis of
flight telemetry data and space communications.

~

APPLIED MATHEMATICIANS are sought with strong analytical abilities, extensive
knowledge of advanced techniques in numerical analysis for computers, and experience
in mathematical investigations related to advanced engineering programs. An MS or PhD
in mathematics or theoretical physics is necessary.

_ - - - - - - - - - - - - - - - - - Diversified Pos~tions for - - - - - - - - - - - - - - - - -......
SENIOR COMPUTER PROGRAMMERS
ANALOG COMPUTER PROGRAMMERS
As Senior Programmers at the Missile and Space VeBSEE, Physics or math degree required. Will plan
hicle Department you will have all the advantages of
sequence of computer operation, determine the
an extensive computer facility. An IBM 704 is currently
circuitry for engineering problems, set up and
operate computer.
in use; a 7090 is due for installation in 1960. The work
covers analysis and programming for technical data
The work is in a growing analog facility which
systems, flight test data systems and advanced space
includes Electronic Associates and Reeves
programs. Requirements include ability to direct junior
Analog Equipment, a combined Analog-Digital
programmers, a BS or advanced degree, minimum of
Facility and a passive element analog computer.
2 years experience on a large scale, binary computer.
For further information regarding
opportunities here, write Mr. Thomas H. Sebring, Div. 56 21D
You mill receive an answer within 10 dO!1s.
MISSILE & SPACE VEHICLE DEPARTMENT

GENERAL.ELECTRIC
319a Chestnut Street, Philadelphia 4, Pa.

,

COMPUTERS and AUTOMATION for May, 1960

,

'

.'"

., ,.

•

~

:

..

,"

.. •

•

I-

•

17

R

ea

d ers'and,

Editor s Foru

FRONT COVER: RAPID INDEXING OF
THOUSANDS OF CHEMICAL ARTICLES
The front cover shows the actual printing out of
two pages of a new journal called "Chemical Titles,"
which is to be produced entirely by computer, once the
initial titles and authors of articles have been punched
on punch cards. The publisher of this journal is the
American Chemical Society, 2 Park Ave., N ew York
16, N.Y.; the machine which prints the master pages
for photo offset publication is an International Business
Machines Corp. Type 407 Printer; the computer is an
IBM 704; and the method of production is called the
"Keyword-in-Context" method, and has been devised
by H. P. Luhn, of the Advanced Systems Development
Division of IBM, Yorktown Height~, N.Y.
This publication is another milestone in the effort to
inform chemists of what is being published in the field
of chemistry. The journal which does most of the work
in this regard is "Chemical Abstracts" (published by
the American Chemical Society), which in 1960 expects to publish some 150,000 abstracts of papers and
patents, covering 9000 journals representing nearly 100
countries and appearing originally in more than 50
languages. "Chemical Titles" will seek to accomplish
a more modest task but much faster: covering 550 of
the most important chemical journals (including 100
Russian ones) and with a time delay very much less
than the delay of "Chemical Abstracts.." "Chemical
Titles" is in two parts, the keyword-in-context index,
and a bibliography. Each title will appear in the index
as often as it contains a keyword; but 960 words are
rejected as keywords - not only prepositions and conjunctions, but also words such as "use, theory synthesis,
chemistry" which contribute little or nothing to the
recognition of subject matter. Each key word is amplified by the surrounding portion of the title (see the
examples in the front cover picture), and is accompanied by the "identification code" of the article which
the computer determines and which enables the full
title, author, and citation to be quickly found in the
bibliography.
WORKING GROUP FOR BETTER EDUCATION
The first regional meeting of the Working Group for
Better Education was held in a morning and afternoon
session on Saturday, March 26, at Mass. Inst. of Technology, Cambridge, Mass. Nineteen people were present,
3 from Connecticut, 2 frorr. New York, 2 from New
Jersey, 1 from Mexico City (Dr. Sergio Beltran, Director of the Computing Center, National Univ. of Mexico), and the remainder from Massachusetts.
W. Eugene Ferguson, Chairman, Dept. of Mathematics, Newton High School, talked on new developments in secondary mathematics courses and distributed
a guide to them. Norton Levy, Chairman, Dept. of
18

m

Mathematics, Concord High School, talked on the use
of community resources in Concord, consisting of some
80 professional people who have helped in the teaching
of mathematics and in the pursuit of mathematical projects by students. Dr. Robert B. Davis, of Syracuse University, and the School Mathematics Study Group at Yale,
talked briefly on their program. Dr. Sergio Beltran talked
on the thirst for education in Mexico, the lack of means
for fulfilling it, and the tremendous energy going into
the task.
At the end of the morning session, certain resolutions
were discussed and passed unanimously:
-1. That membership in the WGBE should not be
restricted to only members of the Association for Computing Machinery and readers of Computers and Automation, but that membership should be open to all
seriously concerned, academically or technically trained
persons, such as members of the IRE or AlEE or teachers
of mathematics in secondary schools, etc.
- 2. That the WGBE therefore could not ask the
Association for Computing Machinery to pay its costs,
but that it shouJd have membership and dues.
- 3. That dues should be $2 a year, for the present.
- 4. That there be two officers, an Acting Treasurer
and an Acting Secretary. For Acting Treasurer, Frank
Verzuh was elected; for Acting Secretary, Edmund e.
Berkeley was elected.
- 5. That there should be some simple bylaws. For
bylaws committee, the two officers were elected.
- 6. That the WGBE should be a "permissive and
enabling" organization, permitting and helping good
work to get done with a minimum of formality.
It was agreed that six working subgroups should be
started; a list of them (with members as of March 26)
follows:
1A. Teaching computers and data processors to high
school and college students: R. V. Andree, Walter Taylor
Ie. Automatic teaching machines: Frank Verzuh, Ed
Berkeley, Loren Bullock, Robert B. Davis
2e. Logic and reasoning: Werner e. Rheinboldt,
Sheldon Rifkin (both of Syracuse)
3A. Verifying the quality of education produced: Ed
Berkeley
6B. Use of community resources in professional people: Norton Levy
7e. Better education in Mexico: Sergio Beltran, Ed
Berkeley
For the afternoon session, seventeen people reassembled. News of the Los Angeles WGBE group, and of
the Syracuse WGBE group, was reported. For the last
hour and a half an open discussion took place based on
agenda of 16 topics put together by those present.
COMPUTERS and AUTOMATION for May, 1960

THINKING BY MACHINES
1.

From William C. Christian, Asst. Editor
Management and Business Automation
Chicago, Ill.

Management and Business Automation magazine
has tackled the important and exciting subject of the
thinking ability of machines.
The daily press and various magazines are continually
publishing conflicting articles on this subject. The term
"electronic brain" is kicked around and some of the
connotations are actually injurious to the positive progress and development of automation, especially among
users of the equipment.
As an authority in your area of automation, would
you help us clear the air for our readers ? We are going
to make a sincere effort to communicate the full implication of machine thinking between the scientist working
with advanced automation theories and the businessman
who is utilizing the tools of today. We request evidence of a machine's ability to "think" in terms the
businessman can understand. We want to explain how
far today's machines have progressed, and how far they
could go if they were fully utilized. In your opinion:
1. Can or will machines be able to "think" entirely
for themselves?
2. What do you predict the state and extent of automation will be in the next 10 years?
It may be a difficult task to sum up this information in
a brief statement. You may express yourself at length,
if you wish. We would appreciate any information that
would help in the development of this article. In giving
our readers your updated and important opinion, you
will, in effect, be rendering a service, by communicating a vital but much misunderstood message . . . . Your
cooperatioh will be greatly appreciated.
II.

From the Editor

1. The Meaning of "Thinking"

The word "thinking" received most of its meaning
years ago. Its applicability to various kinds of situations
involving reasoning and judgment was established almost entirely in the years before automatic computing
and data processing machines were invented.
As one result, when the word "thinking" is applied to
machines, the connotation of the word "thinking"
arouses emotional reactions in people.
As a second result, there is an uncertain and fuzzy
area in the meaning of the word "thinking" when it is
applied to novel situations. The operation of automatic
machines that handle information reasonably and automatically is a novel situation.
Therefore, most of the argument about whether or
not a machine can think is shallow and fruitless, because it reduces to an empty argument over whether or
not some kind of specific activity (for example, like
carrying on a sensible conversation) is or is not to be
included under the term thinking. It does not seem to
be worthwhile to be much interested in this part of the
argument any longer, since the answer in any discussion
depends upon which person in that discussion succeeds
in capturing the right to define the word "thinking."
COMPUTERS and AUTOMATION for May, 1960

2.

Education and Thinking

A more important part of the argument is based on
considering how a human being manages to think.
Clearly, he does not succeed in thinking adequately in
the way human beings think unless he has been educated, and rather well educateci at that. A completely
uneducated human being would be one deprived, from
birth on, of any contact with other human beings; such
a person cannot think, cannot even communicate.
Let us contrast briefly the education of a human being and a computer, first as to schooling:
Human Being
Computer
Reading
Taking in information
Writing
Putting out information
Spelling
Control over format so that
output is produced in
satisfactory form
Arithmetic, mathematics
Arithmetical unit of the
computer, and some programming
History, geography,
Storage of certain informaliterature, social studies
tion in the memory units
and magnetic tapes of the
computer
Foreign languages
Machine translation
English composition
Capacity to compose sentences having meaning
(shown to be possible in
the report "Conversation
with a Computer" by
L. E. S. Green, published
in the October, 1959,
Computers and Automation)
Let us compare some other subjects which are not part
of the curriculum of formal education in school:
Human Being
Machine
Capacity to look around, Present to some extent 1ll
see, and recognize variradar tracking scopes,
character reading devices,
ous features of the environment
etc.
Capacity to listen to spok- Not present in any existing
en language and extract
machine but being worked on actively
phonemes (unit speech
sounds) from them
Capacity to put phonemes Clearly reducible to a protogether into words
gram for a computer
Capacity to take strings Present in today's compuof words and get an
ters in mathematical and
idea out of them
logical areas
Capacity to take ideas and Present in today's computers
put them into words
Capacity to combine ideas Present in today's machines
in new ways to solve (see for example "A Report
problems
on a General Problem
Solving Program for a
Computer" by A. Newell,
J. C. Shaw, and H. A.
Simon, published in the
July, 1959, Computers
and Automation)
With this list of thinking activities in front of us,
we can ask "Are any human thinking abilities beyond
the capacity of computers?" The answer is clearly "No."
19

We cannot see any insuperable obstacle, any theoretical
barrier. The differences between human being and machine are differences of degree rather than quality.
3.

Personality and Programs

After computers become able to do a ~~al~y ~arg,~
number of the kinds of operations we call thmkmg,
then a computer will appear to have a personalitya way of behaving which is a reflection of the properties of the computer and the way it has been programmed. In the same way, a human being has a personality
which is a reflection of his properties and his education.
But there remains a marked difference. A human being comes as a whole. It is not possible at will to cut
off from him quantities of certain kinds of thinking
behavior, in the way that certain kinds of programs can
be taken out of a computer. Also, it is not possible to
add into a human being, in a minute or two, quantities
of certain other kinds of thinking behavior, in the way
that certain other programs can be put into a computer.
Probably no human being will ever change his personality entirely in one minute, in the way that a progra~
taken out of a computer and another program put m
changes its personality.
Shall we say then the programs do the thinking and
not the computers? (and assert that "the programs are
made by human beings"?)
Nowadays many programs are made by human beings.
But they are mainly transmitting the education with
which they themselves have been programmed by the
processes of more than 5000 years of civilization. But
also computers make programs a~d improve ~rograms;
and this will happen to a much greater extent m the future. For example, the forecast has been made that during the next ten or twenty years the world's champion
chess player will be an automatic computer.
The tide is running strongly in the direction of more
and more thinking abilities becoming parts of the powers of computers. An engineer is well guided when he
says, "If human thinking can do it, then a computer
ought to be able to do it."

COMPUTER BERGERAC
Milton Weiss
Newton, Mass.
In the air-conditioned room overlooking the New
York skyline the lights of the large digital computer
blinked in changing patterns as the machine analyzed
the profits and losses, the sales and earnings of 251
large corporations simultaneously. A battery of teletypewriters connected directly to the computer clattered
noisily at one side of the room, fed questions and financial data to the machine and automatically transmitted
its answers throughout the nation and Canada. Representing the efficient effort of that Stock Exchange firm,
Magill, McCord, Jones and Lindsay to give their branch
offices the best financial and investment advice as quickly as possible, the entire room with its gleaming assemblies of panels and auxiliary operators was in sole
and complete charge of young Dave Harland, who sat
sprawled in a chair before the main panel of the computer.
20

A young man with red hair, lanky legs, big steady
hands, and an MIT degree, Dave knew every transistor,
gear and magnetic scanning head of his computer better
than his own initials. Author of programming shortcuts undreamed of by the computer's maker, he had developed the system of communication with teletypewriters between the branch offices and the machine.
To Dave the computer was more a beloved friend than
a complex maze of pulses, potentials and power, but as
he now gazed moodily at it he didn't see the lights on
the panels nor did he hear the continuous drumming of
the typewriters. Instead his mind was filled with Betty
Marshall. She was more than 200 miles away but her
blue eyes and the smooth curves of her slender figure
were more real to Dave than the shock from a high
voltage power supply.
Suddenly Dave raised his hand and banged the side of
the chair in frustration. For the last week thoughts and
visions of Betty had filled his mind leaving him unable to
concentrate on computers or anything else. Engrossed
with studies and work Dave had never paid any attention
to girls. Besides they always confused him and made
him feel alone and shy. Even trying to talk with a girl
froze Dave's vocal chords and clenched his hands into
tight fists. But last week he had met Betty and more
than anything else he wanted to know her better.
A week ago he had been in Boston to show the teletype operator how to word questions transmitted directly to the New York computer. Betty was seated at the
Boston machine kept by itself for maximum privacy in
a small room. When she smiled up at him while being
introduced, Dave's heart skipped a beat and he gulped.
Somehow he mumbled his pleasure but he could barely
take his eyes from her petite face framed in curly
blonde hair.
He couldn't remember a single word of his instructions to her, but he had no trouble recalling the sound
of Betty's low voice or the occasional touch of her
shoulder as they sat together before her typewriter. Desperately Dave sought for a conversational opening after
the lesson - some way of getting to know her - but
neither the thought nor the words would come. He
returned to New York that same day, a confused and
unhappy young ma.n. And ever since Betty had haunted
him.
And now as he sat in his New York office, Dave was
distinctly unhappy. "This can't go on," Dave reflected
grimly. Abruptly he uncoiled his legs, lifted himself
from the chair, and in two long strides he was beside the
group of teletypewriters. The Boston line wasn't in
use. Good. ~ithout giving himself a chance to think,
he flicked a switch and typed, "Hello, Miss Marshall.
How are you?" Then he typed his name, and waited.
For a long surprised moment the machine remained
silent. Dave held his breath and prayed that Miss Marshall still worked for the company. Then the typewriter
carriage danced jerkily as the reply spelled itself on the
paper. "Fine, Mr. Harland, how are you?"
Dave heard a sigh. With hands that trembled slightly he responded that he too felt fine and that he hoped
she was having no trouble with her machine. Miss Marshall replied that all was well and that his instructions
of last week had proved very helpful.
''I'm very glad to hear it," said Dave, making only two
typing errors. Then he stopped. His fingers remained
COMPUTERS and AUTOMATION for May, 1960

ready on the keys but his brain sent them no message.
Images of Betty, incoherent words, bits of programming
code, all swarmed through his mind, but he could think
of nothing to say - absolutely no way of continuing
the conversation. Dave's scalp tingled and he clenched
his teeth. After a long and awkward pause, he lamely
managed, "Well, see you later."
In Boston, Betty Marshall, making no effort to continue the conversation, typed a quick goodbye. Then
she tore the sheet from the typewriter, and slowly reread it. Very deliberately she folded the paper and put
it in her handbag, a slight smile on her lips.
At the other end of the line, Dave's face was scarlet
as he switched the teletypewriter back to the computer.
He felt like a complete fool. Slowly he turned and walked to the programming console near the computer main
chassis. There he gazed with obvious distaste at the
programming schedules neatly laid out near the large
rolls of magnetic tape.
Then he started to swear, softly and slowly, with deliberate emphasis on each syllable. After a few moments
he stopped, his feelings apparently eased somewhat. He
sat down, shook his head as though to clear it, and
started to program a tape for a new investment analysis
survey.
But now Dave couldn't concentrate on the work.
Thoughts of Betty and his recent performance at the
typewriter interfered with the programming routines.
Several times he found himself punching the wrong coding keys. Finally he made a wry grimace and stopped,
his fingers still on the keyboard. It was no use. Quickly
he erased the tape, and pushed the programming schedule to the back of the desk.
And now that it was too late, thoughts of what he
might have said to Betty flooded through him. Boston
- New York - his visit -- her typing. Each topic presented itself in clear and complete sentences. There was
even a logic in their sequence, a logic which Dave sensed
could almost be programmed on his computer.
Idly Dave pressed the coding keys on the programming board, slowly at first then faster. A smile touched
his lips - no mental block this time. After making sure
that the magnetic tape was in proper position, Dave
worked briskly for about half an hour. Then he flipped
the playback switch. Immediately, the programming
typewriter, actuated by the magnetic tape, pecked out
precise words that paid high tribute to Miss Marshall's
typing ability. There were of course several grammatical
errors, and the spelling was often phonetic. Dave
scanned the words twice, his eyes grave and anxious.
Then he chuckled quietly, turned to the programming
board, and his fingers began to fly.
As though the computer were a trusted friend interceding for him, speaking for him to Betty, Dave's
coded words flowed smoothly onto the tape. Completely
at home with the familiar computer and programming
techniques, Dave at first spoke about light and impersonal subjects - the weather - a play he'd seen - the
crowded city streets. Each topic was to be triggered by
an analysis of words sent by Betty. And Dave's tape
would phrase questions so she would use key words in
answering.
Then as the stream of his thoughts continued, Dave's
talk gradually veered to himself. Cramped over the
desk Dave spoke of his dreams and his hopes, talking to
COMPUTERS and AUTOMATION for May, 1960

that magnetic tape as he could never talk to a girl. It
was near midnight, long after everyone had left the
building, when Dave stood up, his fingers weak and his
legs stiff. For a long moment he gazed at the programming desk with tired eyes. Now he wanted only to
sleep - but he was happier than he'd been for a long
time. He placed the bulky roll of magnetic tape in the
computer then left the office humming under his
breath.
When Dave called Betty late the following afternoon,
her reply came almost immediately. "Hi - how are
things in New York"?
For a moment Dave tried to type his answer in person. His fingers tensed on the keys but again, knowing
he was speaking directly to Betty Marshall, his mind
seemed paralyzed, unable to form even a syllable. He
gritted his teeth but still couldn't think of a word to
say. Even the computer, which had also received Betty's
greeting, seemed to mute its constant hum.
Suddenly Dave's strained expression eased. He swung
away from the typewriter and pressed a new button on
the computer console - a button he had connected early
that morning. Immediately the computer took control
and at a rate approximating Dave's typing speed told
Miss Marshall that things in N ew York were going well
but that memories of Boston were still vivid. The machine then asked Miss Marshall about Boston affairs, and
there ensued a short but lively conversation between
the two. As Dave gazed almost with awe at the typed
sheet unreeling before him, he felt that his work of the
previous night had really been worthwhile.
The computer ended its talk with Betty on a very
cordial note; it even thanked her for her time and
hinted that it was looking forward to speaking with
her again. Dave's face broke into a wide grin and he
patted the machine. Then he scanned the paragraphs
again looking for words or E'hrases that were keys for
new topics. Carefully he tucked the sheets into his
pocket and that night, for the· second time, he worked
late at the programming console.
Dave was hovering over the computer at 4.07 p.m.
several afternoons later when, just as he had arranged,
it again opened the line to Boston and greeted Betty.
He had programmed the computer so that starting at
four 0' clock each afternoon it would check the Boston
line and automatically call Betty when the line was free.
H someone other than Betty replied, the computer
would very politely say the line was being tested and
would then sign off. This arrangement eliminated any
possibility that some one else than Betty, would be on
the line or that Dave might freeze at the teletypewriter
as he had before.
Warm thoughts of Betty entered Dave's mind as he
watched the computer get well into its talk with her.
Wondering idly about new topics to program for her,
he turned and sauntered back to his desk leaving the
computer to its simultaneous tasks of making love to a
girl in Boston while it sent financial data to Texas and
Toronto. Pulling some papers on current analysis toward him Dave sank into his chair. In a few minutes
the image of Betty faded somewhat and the chatter of
the automatic typewriters went unnoticed as he became
engrossed in calculations.
With practiced ease Dave manipulated his slide rule
stopping every so often to make quick computations on
21

a large sheet of paper. Orderly rows of figures covered
most of the sheet when abruptly the tempo of the typewriters changed. Dave glanced toward the machines
and cocked his head. The continuous muffled drumming now seemed to include a rhythmic pounding as
though several keys were being hit again and again in
the same order.
In two strides Dave bounded to the row of typewriters and anxiously looked at each one. With some
relief he saw that all the financial material was flowing
smoothly out over the wires but he stopped short and
gulped when he came to the machine connected to
Betty's line. In marked cadence it was typing the words
- "date with you" - "date with you" - "date with
you" - over and over. The phrases, already spread
over two lines of the roll of yellow typewriter paper,
were rapidly filling a third. That line too was covered
before Dave recovered from his shock and pressed hard
on the button disconnecting the computer from the typewriter.
For a long moment Dave stared down at the typewriter as though it were a snake he had just killed with
a club. Confused and shaken but certain that the computer couldn't be faulty, he closed his eyes and started
a mental search for the coding errors he might have
made in programming the tape.
The sudden banging of the typewriter carriage rudely
shattered his reflection, and the keys started clattering a
message from Betty. "What's going on? Are you all
right, Dave?"
A flood of red climbed Dave's face and he gripped the
typewriter table trying to think of a reply. His mind
awhirl, there was no way of explaining the breakdown
to Betty. Besides the thought of addressing her himself
started to overwhelm him again. Biting his lips Dave
forced stiff fingers onto the typewriter keys and pecked
out "trouble on line - sorry - see you later." Without waiting for a reply he abruptly switched off the
circuit and dashing back to the computer he viciously
yanked out the offending magnetic tape.
H was almost three the next morning before Dave replaced the tape, corrected and with many additions,
back in the computer.
Circles rimmed his eyes when Dave slowly walked
into his office the next morning at nine o'clock. Mr.
Robert Lindsay was seated at the computer console waiting for him.
"Good morning," Dave said surprised. It wasn't often that a partner of the firm visited him, and none had
ever come in that early in the morning to the computer
room. Cautiously he searched Lindsay's face. Did he
know about Betty, he wondered.
"Morning, Dave," said Lindsay. His steady blue
eyes smiled without even a hint of disapproval. "Don't
take off your coat," he cautioned. ''I've got plane tickets
to Boston."
"What for?" asked Dave. He tried to keep his voice
casual.
"The Boston office just called. Some English investment men want to hear about your computer before they
leave for home today." Lindsay smoothed his iron-gray
hair and reached for his hat. "It may mean more business."
"What about the computer?" protested Dave.
22

"It's automatic, Isn t it?" said Lindsay. ''I'll have
some one look in here during the day." He started walking toward the door, beckoning. "Besides, Boston said
their line wasn't quite right. Perhaps you ought to look
at it." His tone was even and impersonal.
Dave swallowed and glanced sharply at Lindsay, then
followed him out of the room.
Once in Boston Dave was keenly disappointed to find
that the meeting was being held in a hotel rather than
the office where Betty worked. The English visitors had
seen the teletype responding to the computer, and now
they wanted information on the computer itself before
leaving for London. Surprisingly they were interested
in the technical details of the machine as well as its cost
and abilities. Completely at ease on matters concerning
the computer he knew so well, Dave described it at
length, and it was late afternoon before he had answered
all the questions and received admiring thanks.
The meeting over, Dave excused himself, and leaving
Mr. Robert Lindsay to see the visitors to their plane, he
hurried to the Boston office. He was tired as he started
the walk to Federal Street but the anticipation of seeing
Betty soon put bounce in his step and made his temples
throb with an increasing excitement.
Once at the office, he barely stopped to greet some of
the men he knew, then hurried to Betty's tiny room.
He stood quietly in the doorway for a moment, drinking her in with his eyes, his face calm, his heart pounding. Her back toward him, Betty sat erect in the chair,
her slim shoulders moving slightly as her fingers travelled over the teletypewriter keys. The late sun coming
through the window turned her hair golden.
Finally Betty stopped typing. Only then did Dave
move. Quietly he stepped into the room and coughed.
Hearing him Betty turned in her chair and looked at
him. For a fleeting moment she stared without recognition and then her eyes widened in surprise.
"Dave," she exclaimed, "what are you doing here?"
Behind her the keys started to clatter.
Dave managed a small grin but his cheeks started to
warm. "Hi," he said softly, the shyness starting to close
his throat.
Betty smiled and swivelled her chair to face him. Her
sheath-like dress, deep blue in color, outlined the lithe
curves of her young figure and contrasted sharply with
her smooth white skin. She opened her mouth to speak
but becoming aware of the typewriter tapping behind
her a puzzled gleam replaced the welcome in her eyes.
Slowly she turned her head to watch the words forming on the yellow roll. Then she looked back at Dave
and very deliberately stood up.
"What sort of a joke is this?" she asked. Her voice
was low but her eyes were blazing. "If you're Dave
Harland, who is on the other end of that line?" Angrily
she pointed to the machine steadily typing its message.
Dave's muscles tensed and abruptly he snapped his
head toward the machine. In a single stride he was beside it reading the words. He grunted as though a fist
had hit him in the belly. The computer was trying to
persuade Betty to visit N ew York and see the sights
with him. It was making conversation just as Dave had
ordered it to.
"I forgot to take out the tape," he said weakly.
COMPUTERS and AUTOMATION for May, 1960

"That's me - I mean - the computer." He took a step
toward Betty and tried to smile.
Betty retreated toward the doorway. "What are you
talking about?" she asked. "Is this some game you and
your pals are playing?"
Dave's face was crimson. "Oh, no!" he protested.
You see - I couldn't -". And then he choked. The
words to tell her that a computer had been programmed
to speak for him simply would not form. Yet as he
looked at Betty, standing stiff and angry near the doorway, he longed for her more than ever.
"Wait," he said hoarsely, "Wait, let me show you!"
Then he turned to the machine and typed a single
line. "Do you like me?" Behind him he heard Betty
approach. Without looking at her he said, "The computer thinks you sent that."
"Y ou must be insane" Betty said, but she remained,
staying a stiff foot distant from Dave.
Together they saw the letters form a reply, ''I'm falling in love with you."
"It's a trick," Betty said blushing. "No machine can
do that."
To Dave this criticism of his machine and his ability
had to be answered. Forgetting his shyness he blurted,
"Would anybody repeat the same phrase over and over
like the computer did yesterday?"
"Maybe not," retorted Betty, "but why use the computer at all? Can't you speak up for yourself?"
Dave looked at Betty soberly, then he smiled. "It's
much easier to talk to the computer than to you," he
confessed. "Watch."
Turning back to the machine he typed out Betty's
question. "Why do you use the computer? Why don't
you speak for yourself?"
There was a slight pause and then the computer under the control of Dave's tape replied - "Some men
talk well, others write, or paint, or sing. The genius of
each is stimulated by love, and each best makes love in
the language of his genius. Computers are my language."
Betty opened her lips to speak but Dave shook his
head and pointed at the machine. The computer was
telling of Dave's lonely shyness and how hard it had
been to meet and talk to girls. It told how he had tried
to talk to her the first time on the teletypewriter and
failed, and how he had poured out his heart to the magnetic tape program, and finally some of the way he had
programmed the system to talk with her.
After an outpouring of teletyped words, the machine
stopped. Betty walked over to the window and looked out
into the dimming light at the crowds of people going
home. Dave stayed quiet but his eyes followed her. In
the stillness he became conscious of the traffic noisily
flowing past.
Finally Betty turned and walked over to him. "Why
not have dinner at my house tonight, Dave," she said.
He saw that her eyes were moist but that she was smiling.
"Dad's an electronics engineer. He'd like to hear about
your computer."
Dave started to put his arms around her but Betty
gently pushed him away and touched her finger to his
lips. "There are some words you must first learn to
speak for yourself, Dave." Then she added, musingly,
"Perhaps we can learn them together."
COMPUTERS and AUTOMATION for May, 1960

CLARE ED Sealed Contact Reed relays put you ... the
designer ... in the driver's seat. They are simple in
design, flexible in assembly. They are packaged and
mounted to comply with your mechanical design configuration ... even on your own circuit board.
CLAREED relays are ideal components for transistordrive applications, computers, data-processing and
other high speed equipment.
Contacts are hermetically sealed in inert gas. Tens
of millions of operations are assured since contact
contamination is completely precluded. Hundreds
of millions of operations are possible when operated
up to % rated load.

ACTUAL SIZE / Basic element of CLARE ED relays is
this switch capsule. A pair of magnetically operated
contacts is hermetically sealed in a glass capsule, in
an atmosphere of inert gas. The capsule combines
extreme simplicity and long life. It has excellent
low-level characteristics.
If you use relays, it will pay you to know all
about CLAREED ••• an entirely new concept in
relay design. Address C. P. Clare & Co., 3101
Pratt Blvd., Chicago 45, Illinois. In Canada:
C. P. Clare Canada Limited, Box 134, Downsview, Ontario • Cable Address: CLARE LAY.
Send for Bulletin Cpe-5.

When only the best Is good enough

23

BOOKS
and
OTHER PUBLICATIONS
Moses M. Berlin
Cambridge, Mass.

E PUBLISH HERE citations
and brief reviews of books
W
and other publications which have a
significant relation to computers, data
processing, and automation, and
which have come to our attention.
We shall be glad to report other information in future lists if a review
copy is sent to us. The plan of each
entry is: author or editor / title /
publisher or issuer / date, publication
process, number of pages, price or
its equivalent / comments. If you
write to a publisher or issuer, we
would appreciate your mentioning
Computers and Automation.
Data Processing Seminar on Status of
Digital Computer and Data Processing
Developments jn the Soviet Union /
Office of Naval Reseal'ch, Dept. of the
Navy, Washington, D.C. / 1958, offset,
179 pp, cost ?
Four computer people, specialists in
the field of digital data processing, offer
reports on progress in the field in the
Soviet Union, having visited a number
of compl!lter installations there, during
the summer of 195,8,
John W. Carr III of the University of
Michigan covers tours, talks, physical
structure, programming theory, mathematical techniques, logic, and, in two
appendices, key excerpts from Soviet literature pertinent to the report. A bibliography on Soviet Computer literature
is included. A. J. Pedis, Carnegie Institute of Technology, reports on the comparatvie merit of Russian coding systems. Excerpts from Computer publications are included; in the second of two
appendices, the internal design of the
Soviet computer is described.
The report by James E. Robertson,
University of Illinois, elaborates on the
design of the computer, and the report
by Norman R. Scott, University of Michigan, covers the mathematical efficiency
and reliability of Russian data processing
systems.
A discussion session between the four
authors and led by M. C. Yovits, ONR,
follows the reports.
An Annotated Bibliography, Univac Educational Series No. 3 / Remington
Rand Univac, 315 Fourth Ave., New
York, N.Y. / 1959, printed, 54 pp,
cost ?
24

This bibliography of 267 references
has been prepared for those whose training in computers has been either limited or extensive. This second edition
comprehensively covers: the history of
digital computers; books on theory and
operation; applications; and general background and sources. The last section incl udes lists of indexes, periodicals, bibliographies, and glossaries.

The Systems Approach to Electronic Wiring and Connections / Methode Manufacturing Corp., 7447 W. Wilson Ave.,
Chicago 31, Ill. / 1959, printed, 28 pp,
free on request
This pamphlet discusses many possible
ideas for wiring device advancement. In
three sections, "case histories" of printed
circuit and wiring device applications
are described, and check lists to assist in
specification are given.

Report of the Commission on Mathematics, Appendices / College Entrance
Examination Board, c/o Educational
Testing Service, Box 592, Princeton,
N.J. / 1959, printed, 231 pp, $1.00
The purpose of this edition of appendices is to provide teachers with information, instruction and enrichment, to
help the teacher implement the recommendations for a more advanced highschool curriculum. Algebra, geometry and
trigonometry, are covered, with an index following the text.

Automation, Annual Product Review /
Penton Pub. Co., Penton Bldg., Cleveland 13, Ohio / 1959, printed, 35 pp,
cost?
Precisely what the title states, a catalogue of products, from actuators to
worm gear jacks, with computers included. The contents consist of descriptions of automation equipment, products, components, and services advertised
in Automation during the past twelve
months.

Goldman, Stanford, Dept. of Elec. Engrg,
Syracuse Univ., Research Inst., Syracuse, N.Y. / Cybernetic Aspects of
Homeostasis (PB 151733) / Office of
Technical Services, U.S. Dept. of Commerce, Washington, D.C. / 1958, offset,
67 pp, $2.00
"The maintenance of a prescribed internal environment in the body of an
organism, in spite of wide fluctuations
in its activity and in its external environment, is called homeostasis." This paper
discusses the principles of feedback,
servo, and information theory, and describes their relationship with homeostasis, and what light these subjects throw
on homeostasis.
Murphey, Robert W. / How and Where
to Look It Up / McGraw-Hill Book
Co., Inc., 330 West 42 St., New York
36, N.Y. / 1959, printed, 721 pp,
$15.00
This reference book provides a comprehensive guide to standard reference
sources, containing more than 3,900
topics of interest for research. An introduction explains how to make efficient
use of the book, followed by three sections discussing: reference works and
their use, basic types of reference, and
specific sources of information. Encyclopedias, dictionaries, periodicals, almanacs, handbooks, biographical works
and foreign informational sources are included. An alphabetical subject heading
list, and an index, are included.

Annales de l' Association Internationale
pour Ie Calcul Analogique, for September, 1959, vol. 1 no. 6 / Presses
Academiques Europeennes, 98, chaussee de Charleroi, Bruxelles 6, Belgium
/ 1959, printed, 64 pp, cost ?
This issue is published mainly in
French, with certain articles in English,
it includes abstracts of articles, a bibliography, items about the European
computer scene. The articles in English
are: "A Special Function Generator: the
Frequency Programming System for the
25 GeV C. E. R. N. Proton Synchroton,"
and "Digital vs. Analog Computation."
LGP-30 ACT 1 Compiler / Royal McBee
Corp., Data Processing Div., Port
Chester, N.Y. / 1959, offset, 18 pp,
free on request
A detailed description of an Algebraic
Compiler and Translator, which has been
designed to facilitate the coding of problems through the use of common mathematical terminology.
The technique
makes use of basic algebraic rules to
eliminate the need for many involved
sets of instructions. This edition includes
definitions of terms, examples, and in
four appendices, a scaling technique,
brackets and precedence, subroutine and
calling sequence, and assembly for ACT
1.

Dummer, G. W. A. / Modern Electronic
Components / Philosophical Library,
15 E 40 St., New York 16, N.Y. /
1959, printed, 472 pp, $15.00
A reasonably comprehensive summary

COMPUTERS and AUTOMATION for May, 1960

of essential data on electronic components. Contains thirty-two chapters on:
component specifications and characteristics, reliability, behaviour under arduous
environmental conditions, and testing
methods and techniques. Also includes
a brief history of component development
in Great Britain, and an index.
Proceedings of the Eastern Joint Computer Conference, Boston, Mass., Dec.
1-3, 1959 / Eastern Joint Computer
Conference, Nat'l Joint Computer Committee, IRE, One East 79 St., New York
21, N.Y. / 1959, printed, 260 pp, $3.00.
This volume contains thirty-two papers
which were delivered at the conference.
A number of the new data processing
systems are described-some, in great
detail. Inc Iud e dare S T RET C H ,
Mobidic, SIMCOM, and Univac-Lare.
Other papers discuss methods for solving
mathematical problems on a computer,
applications of Boolean algebra, etc.
Computer Language Translator-Application Information / Electronic Engineering Co. of Calif., 1601 East Chestnut
Ave., Santa Ana, Calif. / June, 1959,
printed, 20 pp, free.
The Computer Language Translator
provides for rapid and efficient translatlOn between different types of automatic
data processing systems. This publication
provides information on: translation capabilities; the basic translation system; five
common translation modes; and physical
characteristics of the translator. An appendix presents systems delivered or in
production.
Caldwell, W. I., G. A. Coon and L. M.
Zoss / Frequency Response for Process
Control/McGraw-Hill Pub. Co., Inc.,
330 West 42 St., New York 36, N.Y. /
1959, printed, 395 pp, $11.50.
The fundamental methods of frequency
response and their application to the
analysis, testing, and design of process
control systems, compose the two parts
of this book. In the "Basic Theory" section, certain methods of analysis are given,
with a careful treatment of distancevelocity lag, linear lag, and stability considerations.
These methods-which reduce the
amount of mathematical computation usually needed-are applied to problems.
Included are techniques for calculating
system time constants, and material on
the dynamics of temperature measurement. A chapter of the book discusses a
technique for converting from step to
frequency response, using a desk calculator. TwO' appendices include a list of
symbols, and construction of templates.
Value Engineering, 1959 / Engineering
Publishers, Elizabeth, New Jersey /
1959, printed, 165 pp, $6.00.
This book contains the complete versions of the technical papers presented
at the Electronic Industries Association
Conference on Value Engineering, Oct.

5-6, at the Univ. of Penn. The nineteen
papers discuss this new type of engineering, and its use by a large number of
companies.
Kirchmayer, Leon K. / Economic Control
of Interconnected Systems / John Wiley
& Sons, Inc., 440 Fourth Ave., New
York 16, N.Y. / 1959, printed, 207 pp,
$12.50.
This book presents for electrical engineers theories and applications for
optimizing computer control in a continuous process industry including power
systems (steam and hydro). In eight
chapters, the author discusses: the use of
mathematical methods to determine equations whose solution would result in
optimum economic performance; the use
of theoretical and differential analyzer
methods of predicting the dynamic performance of interconnected systems; and
the integration of computer and controller
to obtain a computer-controlled system.
An index is included.

Goodman, L. L. / Automation Today and
Tomorrow / Essential Books, 1600
Pollit Drive, Fair Lawn, N.J. / 1959,
photooffset, 158 pp, $7.20.
From a purely technical point of view,
this is an up-to-date report on the progress of automation. The first part of
the book discusses computer control
and automation in such fourteen varied
industries, including machine tools, sugar
beets, chemicals, pulp and paper, ice
cream, and concrete. The second part
presents an extensive bibliography including informative books on most of the
technical advances in automation. Subject
and author indexes follow the bibliography.
Automation Systems-Proceedings of the
Electronic Industries Association Conference on Automation Systems, 1958 /
Engineering Publishers, GPO Box 1151,
New York 1, N.Y. / 1958, printed,
180 pp, $5.00.
Seventeen papers delivered at the
Conference are presented, including: "The
Systems Concept as Applied to Automation"; "Reliability of Production Equipment Testing"; 'Progress Toward Nationwide Integrated Data Processing Systems";
and "The Interest of the Federal Government in Automation." Topics covered:
Automation Within the Electronics
Industry, (two sessions); Automation
Outside the Electronics Industry, (two
sessions) ; The Economic, Educational and
Social Aspects of Automation; a Summation session.

Zemanek, Heinz / Elementare Informations theorie /
R.
Oldenbourg,
Munchen, den Rosenheimer Strasse 145,
W. Germany / 1959, 120 pp, cost?
A German-language introduction to
information theory, with five chapters
and two appendices which include tables
and lists of functions used in the text.

COMPUTERS and AUTOMATION for May, 1960

Oppelt, W. / Anwending von Rechenmaschinen bei der Berechnung von
Regelvorgangen / R. Oldenbourg,
Munchen, den Rosenheimer Strasse 145,
W. Germany / 1959, printed, 128 pp,
cost?
A German-language discussion of
analog-digital computing systems. Numerous diagram accompany the text,
which describes the logical design and
application of the machines.
Bowker, Albert H. and Gerald J. Lieberman / Engineering Statistics / PrenticeHall, Inc., 70 Fifth Ave., New York 11,
N.Y. / 1959, printed, 585 pp, $11.00.
This book is intended as a text for a
first course in statistics, for students in
engineering and the physical sciences.
There are thirteen chapters with a great
number of sections on probability, random
variables, distribution theory, estimation,
analysis of variance, significance tests,
analysis of enumeration data, statistical
quality control, and sampling inspection.
A comprehensive treatment on how to
fit straight lines is offered. An appendix
contains various tables relevant to the
study; and index is included.

Data Processing-1959 Proceedings /
Nat'!. Machine Accountants Assn.,
Mt. Prospect, III. / 1959, printed, 394
pp, $10.00.
The general theme of the forty-six
papers presented in this book, is the
development and applicatlOIl of data processing in the past year. Among the
topics are: "Why Automate?", "Planning
New Applications," "Progress in the
Data Processing Industry," "On Job
Training," "Determining E qui p men t
Needs," ete. The book, which lacks an
index, contains papers which were given
June, 1959, at the National Convention
of the NMAA ;some have been published
in the "Journal of Machine Accounting."
Proceedings of the Fourth National Conference of Bendix G-15 Users Exchange
Organization / Bendix Aviation Co.,
Computer Div., Calif. / 1959, offset,
no. of pp not stated, limited distribution.
A collection of the reports delivered
at the conference, is presented. No table
of contents; no consecutive paging.
Perry, J. W., and Allen Kent / Tools
for Machine Literature Searching /
Interscience Publishers, Inc., 250 Fifth
Ave., New York 1, N.Y. / 1958,
printed, 972 pp, $27.50.
This volume provides, in four parts:
Introduction to
Machine
Literature
Searching; Engineering of Machine Literature Searching Systems; Procedures for
Analyzing, Encoding and Searching of
Recorded Information; and a Thesaurus
of Scientific and Technical terms. The
last part includes the Semantic Code dictionary, useful as a cross reference tool
and as an authority for developing subject heading lists.
25

CALENDAR OF COMING EVENTS
May 3-5, 1960: Western Joint Computer Conference, Jack
Tar Hotel, San Francisco, Calif.
May 9-11, 1960: Meeting, Burroughs 220 Computer
User Group (CUE), San Francisco, Calif.; contact
'Merle D. Courson, First National Bank of San Jose,
San Jose, Calif.
May 9-12, 1960: 2nd ISA Instrument-Automation Conference and Exhibit of 1960, Civic Auditorium and Brooks
Hall, San Francisco, Calif.
May 17-18, 1960: Symposium on Superconductive Techniques for Computing Systems, sponsored by Information Systems Branch, Office of Naval Research, at
Dept. of Interior Auditorium, Washington, D.C
May 18-20, 1960: Operations Research Society of America
Seventeenth National Meeting (Eighth Annual Meeting), Statler-Hilton Hotel, New York, N.Y.
May 19, 1960: Conference on Parallel Programming, sponsored by Cleveland-Akron Chapter of Association for
Computing Machinery, Cleveland, Ohio / Contact: 1. R.
Turner, NASA, Lewis Research Ctr., 21000 Brookpark
Rd., Cleveland 35, Ohio
May 23-25, 1960: 9th Annual Telemetering Conference
(West Coast), sponsored by ISA with ARS, AlEE
and ISA cooperating, Miramar Hotel, Santa Barbara,
Calif.
May 24-27, 1960: Conference on Automatic Computing
and Data Processing in Australia, Universities of Sydney
and New South Wales, Sydney, Australia.
June 1-3, 1960: 6th Annual ISA Instrumental Methods
of Analysis Symposium, Montreal, Canada
June 2-4, 1960: National Conference of University Computing Center Directors, Shoreland Hotel, Chicago, Ill.;
contact Mrs. Robert Drew-Bear, American Mathematical
Society, 190 Hope St., Providence 6, R.I.
June 6-7, 1960: Second Conference of The Computing
and Data Processing Society of Canada, University of
Toronto.
June 13-24, 1960: University of Michigan Engineering
Summer Conferences, Courses in Computer Science and
Engineering, University of Michigan, Ann Arbor, Mich.
June 14-17, 1960: Seminar on Systems Simulation Using
Digital Computers, a Cornell University Industrial Engineering Seminar, Cornell University, Ithaca, N.Y.;
contact J. W. Gavett, Seminars Coordinator, Dept.
of Industrial and Engineering Admn., Upson Hall,
Cornell Univ., Ithaca, N.Y.
June 15-29, 1960: 7th Rassegna International Electronics,
Nuclear Energy, and Cinematography Scientific Congresses and Exhibition (included in the program: Electronic computers - collation and processing of data for
research operation), Palazzo dei Congressi, Rome, Italy.
June 22-24, 1960: 1960 National Conference and Business Show, National Machine Accountants Association, Mark Hopkins and Fairmont Hotels, and Calif.
Masonic Memorial Temple, San Francisco, Calif.
June 25 - July 5, 1960: 1st International Congress for
Automatic Control, AACC sponsored, with ISA,
ASME, IRE and AICE cooperating, Moscow, U.S.S.R.
26

July 19-22, 1960: USE Organization Meeting, Syracuse,
N.Y.; contact James W. Nickitas, 315 Park Ave .. New
York, N.Y.
Aug. 6-7, 1960: 3rd Annual Conference of the Northwest
Computing Association, Portland, Ore.
Aug. 8-12, 1960: Pacific General Meeting of the American
Institute of Electrical Engineers, San Diego, Calif.
Aug. 10-12, 1960: Annual Meeting G-15 Users' Exchange
Organization, Pittsburgh Hilton Hotel, Pittsburgh, Pa.;
contact Dr. Jerry C. L. Chang, Richardson, Gordon &
Associates, 3 Gateway Center, Pittsburgh 22, Pa.
Aug. 23-26, 1960: WESCON, Ambassador Hotel and Pan
Pacific Auditorium, Los Angeles, Calif.
August 23-25, 1960: Annual Meeting of the Association for Computing Machinery, Marquette Univ.,
Milwaukee, Wisc.
Sept. 19-21, 1960: 5th Annual Symposium on Space Electronics and Telemetry, sponsored by The Institute of
Radio Engineers, Inc., Shoreham Hotel, Washington, D.C
Sept. 20-24, 1960: Symposium on the Numerical Treatment of Ordinary Differential, Integral and IntegroDifferential Equations, University of Rome, Rome, Italy;
contact Prof. A. Ghizzetti, Provisional International
Computation Ctr., Palazzo degli Uffici, Zona dell'EUR,
Rome.
Sept. 22-23, 1960: Fall Meeting of The Univac Users Association, Washington, D.C; contact W. C Rockwell,
Remington Rand, 315 Park Ave. So., New York 10, N.Y.
Sept. 26-30, 1960: 3rd ISA Instrument-Automation Conference and Exhibit of 1960, and ISA's 15th Annual
Meeting, New York Coliseum, New York, N.Y.
Oct. 4-6, 1960: Meeting, Burroughs 220 Computer User
Group (CUE), Philadelphia, Pa.; contact Merle D.
Courson, First National Bank of San Jose, San Jose,
Calif.
Ocr. 9-14, 1960: 1960 Fall General Meeting of American
Institute of Electrical Engineers, New York, N.Y.;
contact Clarke S. Dilkes, Assoc. Dir., Burroughs
Corp., Research Ctr., Paoli, Pa.
Ocr. 10-12, 1960: National Electronics Conference, Hotel
Sherman, Chicago, Ill.; contact Prof. Thomas F. Jones,
Jr., NEC Program Chairman, School of Electrical
Engrg., Purdue Univ., Lafayette, Ind.
Oct. 17-19, 1960: Symposium on Adaptive Control Systems, sponsored by Long Island Section, Institute of
Radio Engineers, Garden City Hotel, Garden City,
L.I., N.Y.; contact F. P. Caruthers, Symposium Chairman, c/o Specialties Inc., Skunks Misery Rd., Syosset,
N.Y.
Nov. ?, 1960: 13th Annual Conference on Electronic
Techniques in Medicine & Biology, sponsored by ISA,
with IRE and AlEE cooperating, Washington, D.C
Sept. 6-8, 1961: 1961 Annual Meeting of the Association
for Computing Machinery, Statler Hotel, Los Angeles,
Calif.; contact Benjamin Handy, Chairman local Arrangements Committee, Litton Industries, Inc., 11728
W. Olympic Blvd., W. Los Angeles, Calif.
Dec., 1960: Eastern Joint Computer Conference, New
Yorker Hotel, New York City; contact Dr. Nathaniel
Rochester, IBM, Yorktown Heights, N.Y.
COMPUTERS and AUTOMATION for May, 1960

SURVEY
OF
RECENT
ARTICLES
Moses M. Berlin
Cambridge, Mass.

We publish here a survey of articles related to computers and data
processors, and their applications and
implications, occurring in certain
magazines. We seek to cover at least
the following magazines:
Automatic Control
Automation
Automation and Automatic
Equipment News (British)
Business Week
Control Engineering
Datamation
Electronic Design
Electronics
Harvard Business Review
Industrial Research
Instruments and Control
Systems
ISA Journal
Proceedings of the IRE
The Office
Scientific American
The purpose of this type of reference information is to help anybody
interested in computers find articles
of particular relation to this field in
these magazines.
For each article, we shall publish:
the title of the article / the name of
the author(s) / the magazine and
issue where it appears / the publisher's name and address / two or
three sentences telling what the
article is about.
Office Automation - A Challenge to Personnel Relations / Raymond Dreyfack
/ Punched Card Data Processing, vol.
1, no. 7, Nov.-Dec., 1959, p 15 / Gille
Associates, Inc., 956 Maccabees Bldg.,
Detroit 2, Mich.
This article discusses the personnel
problems created by a change to automation. Allaying the fears of employees is a
problem which must be dealt with honestly, before and during the implementation
of automatic data processing.
A Comparison of Integral [Transfer] and
Incremental [Transfer] Digital Computers for Process Control Applications
/ Edward 1. Braun / Control Engineering, vol. 7, no. 1, Jan., 1960, p 113
/ McGraw-Hill Publg. Co., Inc., 330
West 42 St., New York 36, N. Y.

"Integral transfer computers" are digital computers which are general purpose
computers and are not digital differential
analyzers. "Incremental transfer computers" are digital differential analyzers.
Each type of computer is discussed in regard to its suitability for process control
applications, and the general conclusion
in regard to which to use appears to be
"it depends."
Computer Uses / Instruments & Control
Systems, vol. 32, no. 12, Dec., 1959,
p 1816 / Instruments Publg. Co., Inc.,
845 Ridge Ave., Pittsburgh 12, Penna.
Thirteen computer applications are listed and briefly described. Among the systems which are computer controlled: electric power station polymerization process,
electric power generation, and natural gas
liquids cracking.
Performance of a Solid-State Process
Computer / Joseph A. Reine, Jr. /
Instruments and Control Systems, vol.
32, no. 12, Dec., 1959, p 1820 / Instruments Publg. Co., Inc., 845 Ridge Ave.,
Pittsburgh 12, Penna.
This article presents the performance
record of an installation in a power and
light company, which includes a solidstate logger computer. Equipment performance, maintenance and trouble shooting activities are described.
Computer Control Philosophy / Clifford
E. Mathewson and Byron White / Instrument and Control Systems, vol. 32,
no. 12, Dec., 1959, p 1830 / Instruments Pub. Co., Inc., 845 Ridge Ave.,
Pittsburgh 12, Penna.
The two authors diSCUSS computer control in side-by-side articles, presenting information on computers designed for specific applications, analog and digital systems, mathematical model uses, and control by automatic experimentation with
the process.
A Computer Pays Off in Plant Research /
J. A. Curran / Automation Progress,
vol. 4, no. 9, Sept., 1959, p 300 / Leonard Hill Technical Group, Leonard Hill
House, Eden St., London, N. W. 1.
Appreciable savings are realized in a
heat transfer process which is computer
controlled. The research work is being
performed in tests of heat transfer in a
full-scale heat exchanger pilot plant; the
computer's functions are described.
Some Case Histories of Business Computers in the U. S. A. / Journal of Institution of Electrical Engineers, June,
1959, p 347 / J. of I. of E. E., Savoy
Place, London WC 2, England.
This article describe..; computer installations of four U. S. companies, including a bank, a major airline, and a radio
company. Results from the data processing systems are given, with a number of
suggestions for successful implementation
of computer use.
Compact Memories Have Flexible Capacities / D. Haagens, Engineering
Mgr., General Ceramics Corp., Keasby,
N. J. / Electronics, vol. 32, no. 40,
Oct. 2, 1959, P 50 / McGraw-Hill Pub.
Co., Inc., 330 West 42 St., New York

.,6,N. Y.
A description of a technique - coincident-current - for constructing digital
computer buffer memories. The system

COMPUTERS and AUTOMATION for May, 1960

•.. that possesses the mathematical logic
of giant computers, yet is easy to operate
and program ... requires no site preparation
or technical personnel ... performs engineering, scientific and business computations
automatically and with electronic speed.
Iterates, compares, branches and handles
sequences of complex operations. If you use
as few as 4 desk calculators for the same
job, the new Clary DE-GO can save you
thousands of dollars every year!
To find out more about this versatile com·
puter, simply mail in this coupon. No obligation, of course.
Franchises available to qualified principals.

r------------l

I

Computer Division
Dept. CA-3
San Gabriel, Calif.
Please send me complete information on
your new electronic
computer, the DE-60.

I
I

I
I
I

I
I
I

I
I
I

I
I Name

I
I

I Company
Position
II Address
Il -City_ _ _ _ _ _Zone
_ _ _State
___

I
II
I

* In

.J

Canada: Computing Devices of Canada, ltd.,
Ottawa~ Ontario

27

,.

wanted:
WAR
GAME
PLAYERS
;

,

Very large-scale air-battle digital
computer simulations are now going on
at the Washington Research Office of
tech/ops. Present operations call for
top-flight mathematicians, mathematical
statisticians, senior programmers,
operations research analysts.

These computer air battles are stochastic
models which involve design and
evaluation, and development of unusual
techniques for studying sensitivity of
these models to input changes. Associated
activity involves design of advanced
programming systems and of common
language carriers which are expected to
be independent of the first computer
used-the computer itself augmenting
and improving the language for use on
later and more sophisticated computers.
A fascinating new book by tech/ops,
THE GAME OF WAR, traces the history
of war gaming from ancient chess of
3,000 years ago to modern computer
gaming, illustrated with authentic
warriors of the periods. For your
free copy, write:

Kingsley S. An dersson

Technical Operations. Incorporated
3S20 PROSPECT STREET, NORTHWEST. WASHINGTON 7, D. C.

28

allows for the addition of plug-in packages to increase capacity, and provides
random and/or sequential access.
Managing Men - Not Machines / L. J.
Hale / Punched Card Data Processing,
vol. 1, no. 7, Nov.-Dec., 1959, p 19 /
Gille Associates, Inc., 956 Maccabees
Bldg., Detroit 2, Mich.
Automation, says the author, will never
replace the need for human beings to
operate and control the automatic controls. This article urges understanding on
the part of management, as they begin to
use computers and data processing. Training of personnel and a knowledge of employees are essential to the success of
any automated industry.
Use of Automatic Computers for Scientific and Technical Calculations and
Direct Production Control / Z. Drab
anll A. Halek / Automatizace, vol. 11,
no. 11, 1959, P 328 / Automatizace,
Praha 2, Spalena ul. 51, Czechoslovakia.
This article, in Czechoslovakian, presents
a biief review of the advantages afforded
by the use of computers for scientific and
technical calculations, and direct production control. Examples are given of possible applications for control in metallurgy, power generation, the chemical industry and transport_
Preparing for an Electronic Data Processing System / Charles J. Graham, First
Nat'I. Bank of Boston / Journal of Machine Accounting, vol. 10, no. 10, Oct.,
1959, P 9 / Nat'l. Machine Accountants
Assn., 720 Kensington Rd., Arlington
Hts., Ill.
This article offers some general solutions and specific illustrations of the
problem of determining the necessity of,
and preparing for, a computer system.
Included are suggestions on recruitment,
selection of equipment, and methods for
efficient application.
European Technical Digests, vol. 4, no.
11, Nov., 1959 / European Technical
Digests, Specialized Information Section, European Productivity Agency,
3 Rue Andre-Pascal, Paris 16, France.
This edition contains a representative
sample of the work accomplished by the
"Digests," including technical improvement papers on: chemicals; plastics, electronics; electrical engineering; food; agriculture; fuel; power; metallurgy; safety;
water; textiles; etc. The publication aims
to publicize industrial process developments on the international scene, and to
cr05sfeed ideas from one national industry to another.
Survey of Progress, and Trend of Development and Use of Automatic Data
Processing in Business and Management Control Systems of the Federal
Government, as of Dec., 1957 / Controller General of the U. S. / Communications of the ACM, vol. 2, no. 5, May,
1959, pp 17-20 / Assn. for Computing
Machinery, Mt. Royal & Guilford Aves.,
Baltimore 2, Md.
This is a report which was delivered to
the Congress of the U. S. The title describes clearly the contents of the report,
which proceeds from early state of development in the computer field, to the need
for research and development, in spite of
great advances over the past decade.
COMPUTERS and AUTOMATION for May, 1960

NEW PATENTS
RAYMOND R. SKOLNICK
Reg. Patent Agent
Ford Inst. Co., Div. of Sperry Rand Corp.
Long Island City 1, New York

HE followin~ i~ a compilation of
patents pertalrung to computers
and associated equipment from the
"Official Gazette of the United
States Patent Office," dates of issue
as indicated. Each entry consists of:
patent number / inventor(s) / assignee / invention. Printed copies
of patents may be obtained from the
U.S. Commissioner of Patents, Washington 25, D.C., at a cost of 25
cents each.

T

November 17, 1959, (Can't):
2,913,178 / Edward J. Petherick, Rowledge, near Farnham, and Geoffrey C.
Rowley, Sutton, Eng. / International
Business Machines Corp., New York,
N.Y. / A coded decimal multiplying
arrangement for a digital computer.
2,913,179 / Bernard M. Gordon, Concord, Mass. I Laboratory f<;>r Electronics, Inc., Boston, Mass. / A synchronized rate multiplier apparatus.
2,913,181 I Jacob Leeder, Aberdeen, Md.
/ U.S.A. as represented by the Sec. of
the Air Force I An electronic scaling
apparatus in analog computers.
2,913,593 / Henry W. Kaufmann, Phoenixville, Pa. / Sperry Rand Corp., a
corp. of Del. I A half-adder for computers.
2,913,594 / John P. Eckert, Jr., Philadelphia, Pa. / Sperry Rand Corp., a corp.
of Del. I A quarter-adder for computers.
2,913,598 / Robert D. Torrey, Philadelphia, Pa. / Sperry Rand Corp., a corp.
of Del. / A transistor core logical
element.
2,913,600 / }. A. Cunningham, Wheaton,
and Arthur W. Holt, Silver Spring,
Md. / U.S.A. as represented by the
Sec. of Commerce / A diode amplifier
and computer circuit.
2,913,705 / Bonnar Cox and Jacob Goldberg, Palo Alto, Calif. I General Electric Co., New York, N.Y. / Apparatus
for transferring digital data from a
first tape to a second tape.
2,913,706 I Ragnar Thorensen and Biagio
F. Ambrosio, Los Angeles, Calif. I
U.S.A. as represented by the Sec. of
Commerce I A transcriber selection
circuit for magnetic drum memory.
2,913,708 / Stephen Paull, Falls Church,
Va. / / A magnetic core nondestructive readout circuit.
November 24, 1959: 2,914,248 / Harold
D. Ross, Poughkeepsie, N.Y., Bernard
1. Sarahan, Bellwood, Ill., and Morton
M. Astrahan, Bennett .Housman, and
Walker H. Thomas, Poughkeepsie,
N.Y. I International Business Machines Corp., New York, N.Y. / A pro-

gram control for a data processing
machine.
2,914,250 I Etienne Honore and Emile
Torcheux, Paris, Fr. I Compagnie Generale de Telegraphie Sans Fil, a corp.
of Fr., and Societe Marocaine de Recherches d'Etudes et de Developpements
"Somarede," a corp. of Morocco / A
function generator for analog computer
systems.
2,914,681 I Floyd G. Steele, La Jolla,
Calif. / Digital Control Systems, Inc.,
La Jolla, Calif. / A logical gating network.
2,914,748 I John R. Anderson, Dayton,
Ohio / Bell Telephone Lab., Inc., New
York, N.Y. I A storage matrix access
circuit.
2,914,751 I William F. Steagall, Merchantville, N.}. I Sperry Rand Corp., a
corp. of Del. I A quarter adder for
computer circuits.
2,914,754 I Karl Ganzhorn and Theodor
Einsele, Sindelfingen, and Hans Bornhauser, Boblingen, Germany / International Business Machines Corp., New
York, N.Y. I A magnetic core memory
system.
December 1, 1959: 2,915,246 / Raymond
G. Piety, Bartlesville, Okla. / Phillips
Petroleum Co., a corp. of Del. I A
polynomial roots computer.
2,915,740 I James B. Ricketts, Jr., Milwaukee, Wis., and Eric E. Bittmann,
Downington, and Joseph Deutsch, Berwyn, Pa. / Burroughs Corp., Detroit,
Mich. I A static magnetic memory system.
December 8, 1959: 2,915,966 I Marvin
Jacoby, Norristown, Pa. I Sperry Rand
Corp., a corp. of Del. I A high speed
printer.
2,915,967 Arthur J. Gehring, Jr., Haddonfield, N.]. and Lawrence F. Harrison, Philadelphia, and Lloyd W. Stowe,
Broomall, Pa. I Sperry Rand Corp., a
corp. of Del. I An information reproducing system.
2,916,209 / Phil A. Adamson, San Gabriel, and Merritt 1. MacKnight, Los
Angeles, Calif. I Hughes Aircraft Co.,
a corp. of Del. I A digital-to-analog
converter.
2,916,210 I Ernst S. Selmer, Oslo, Norway
I Burroughs Corp., Detroit, Mich. I
An apparatus for selectively modifying
program information.
2,916,211 Joachim Schulze, LimbachOberfrohna, and Johannes Hofmann,
Erlangen, Germany I VEB Buchungsmaschinenwerk, Karl-Marx-Stadt, Germany I An electronic calculating machine provided with an arrangement
for rounding off electronic counters.
2,916,727 / Claude E. Jones, Jr., Belleville, N.J. I International Telephone
and Telegrapll Corp., Nutley, N.]. I
A data p-o~ess:ng system.
2,916,728 I Paul R. Gilson, West Covina,
Calif. / Burrou[;hs Corp., Detroit, Mich.
I A magnetic recording and reading
system.
2,916,729 / Stephen Paull, Falls Church,
Va. / - - / A magnetic core binary
circuit.
December 15, 1959: No applicable patents.

COMPUTERS and AUTOMATION for May, 1960

,
,

,

,

and at tech/ops'
Monterey
Research Office, in
California, an unusual
opportunity exists for
a mathematicianprogrammer to expand
his horizons in mathematical analysis and
computer programming,
in an operations research
environment. A B.S.
degree and one to two
years' programming
experience are required,
plus the desire and
ability to contribute to
pre-programming
analysis. The over-all
Company policies in areas
of fringe benefits and
profit-sharing plans are
unique. Write or wire
collect:
Harold E. Kren

Technical
Operations,
Incorporate;~

~

305 WEBSTER STREET
MONTEREY, CALIFORNIA

For your free copy of THE
GAME OF WAR, an informal
history of war gaming from the
ancient Greeks to modern computer gaming, write to Harold E.
Kren, at the above address.

29

December 22, 1959: 2,918,218 I Eric L.
Thomas, Hollywood, and Robert J.
Paul, Belfast, Northern Ireland I Short
Brothers and Harland Lim., Belfast,
Northern Ireland. I An analogue computer.
2,918,574 I Donald]. Gimpel, Chicago,
and William A. Davidson, Evanston,
Ill. I U.S.A. as represented by the Sec.
of the Navy I A digital converter.
2,918,656 I George V. Nolde, Santa Monica, Calif., and Joseph A. Brustman,
Narberth, Pa. I R.C.A., a corp. of Del.
I An information storage apparatus.
2,918,660 I Tung C. Chen, Havertown,
and John H. Lane, Malvern, Pa. I
Burroughs Corp., Detroit, Mich. I A
non-destructive read-out of magnetic
cores.
2,918,661 I Tung C. Chen, Havertown,
and John H. Lane, Malvern, Pa. I Burroughs Corp., Detroit, Mich. I A nondestructive read-out of magnetic memory
elements.
2,918,663 I Tung C. Chen, Havertown, Pa.
I Burroughs Corp., Detroit, Mich. I
A magnetic storage device.
2,918,664 I Edwin W. Bauer, Poughkeepsie, N.Y. I LB.M. Corp., New
York, N.Y. I A magnetic core register.
2,918,669 I Martin L. Klein, Woodland
Hills, Calif. I North American Aviation, Inc. I An arbitrary function generator.
December 29, 1959: 2,919,063 I Donald
R. Young, Poughkeepsie, N.Y. I I.B.M.
Corp., New York, N.Y. I A ferroelectric condenser transfer circuit and accumulator.
2,919,066 I Warren D. White, East Norwich, N.Y. I U.S.A. as represented by
the Sec. of the Army I A vector adding
system.
2,919,354 I Louis A. Russell, Poughkeepsie, N.Y. I I.B.M. Corp., New York,
N.Y. I A magnetic core logical circuit.
2,919,355 I Choang Huang, Ipswich,
Mass. I Sylvania Electric Products, Inc.,
Wilmington, Del. / A bi-stable transistor circuit.

2,919,429 I Francis E. Hamilton, Bing-

hamton, and Ernest S. Hughes, Jr., and
Charles B. Smith, Vestal, and Archie
L. Fun, Endicott, N.Y. I LB.M. Corp.,
New York, N.Y. / A data transfer mechanism.
2,919,430 / Jan A. Rajchman, Princeton,
N.J. I R.C.A., a corp. of Del. I A magnetic switching system.
2,919,431 I Stephen H. Blackford, Endicott, N.Y. / LB.M. Corp., New York,
N.Y. I An apparatus for the magnetic
recording of data.
January 5, 1960: 2,919,857 I Charles R.
Bonnell, Columbia Heights, Minn. I
Minneapolis-Honeywell Regulator Co.,
Minneapolis, Minn. I A torque integrating analog computer with inductive
device.
2,920,193 I Jack Breckman, Haddonfield,
N.]. I U.S.A. as represented by the Sec.
of the Army I A precise analogue store
and impedance transformer.
2,920,313 I David L. Nettleton, Haddonfield, N.J. and Lowell S. Bensky, Levittown, Pa. I R.C.A., a Corp. of Del. I
An information handling device.
2,920,314 I William Miehle, Havertown,
Pa. I Burroughs Corp., Detroit, Mich. /
An input device for applying a synchronously timed data signals to a synchronous system.
2,920,315 I Seymour Markowitz, Los Angeles and Ben T. Goda, Gardena, Calif.
I Telemeter Magnetics, Inc., Los Angeles, Calif. / A magnetic bidirectional
system.
January 12, 1960: 2,920,824 I Walter C.
Lanning, Plainview, N.Y. I Sperry
Rand Corp., a Corp. of Del. / A digital
computer circuit for adding first and
second binary digital numbers respectively.
2,920,825 I Walter C. Lanning, Plainview, N.Y. I Sperry Rand Corp., a
Corp. of Del. / A digital computer for
subtracting from a first binary digital
number a second binary digital number.
2,920,826/ Edward J. Panner, Allentown,
Pa. / Bell Telephone Lab., Inc., New

ADVERTISING

York, N.Y. I An electronic computing
circuit utilizing stepping tubes.
2,920,828 / Billy E. Davis, China Lake,
Calif. / U.S.A. as represented by the
Sec. of the Navy I An analog computer
which permits four quadrant multiplication.
2,921,190 I Franklin H. Fowler, Jr.,
Washington, D.C. I Sperry Rand Corp.,
a Corp. of Del. I A comparison circuit
for detecting coincidence and anti-coincidence of nonoverlapping signals
from two sources.
2,921,192 / Robert F. Casey, Pompton
Plains, and John Gibbon, Morristown,
N.J. I Monroe Calculating Machine
Co., Orange, N.J. I A bistable circuit.
2,921,204 / Donald F. Hastings, Suffern,
N.Y., and John B. Venezia, Leonia, and
Philemon A. Wallace, Bloomfield, N.J.
I Bendix Aviation Corp., Teterboro,
N.J. I A data converter for converting
analog information to digital form.
2,921,296 I Theodore G. Floros, Poughkeepsie, N.Y. I International Business
Machines Corp., New York, N.Y. I A
deskewing device.
2,921,297 I George E. Lund, Havertown,
Pa. I Burroughs Corp., Detroit, Mich.
I A shift code counter.
January 19, 1960: 2,921,737 I Mao Chao
Chen, Palo Alto, Calif. I General Dynamics Corp., Rochester, N.Y. I A magnetic core full adder.
2,921,738 I John P. Greening, Bartlesville, Okla. I Phillips Petroleum Co.,
a Corp. of Del. I A polynomial multiplier.
2,921,740 I Willis E. Dobbins, Manhattan
Beach, Charles R. Williams, Hawthorne,
Hrant H. Sarkissian, Hermosa Beach,
Donald E. Eckdahl and Floyd G. Steele,
Manhattan Beach, and Albert E. Wolfe,
Jr., Compton, Calif. I Northrop Corp.,
Hawthorne, Calif. I A binary incf("
mental slope computer.
2,922,095 I Victor L. Hesse, Playa Dei
Ray, and William O. Fe1sman, Tarzana.
Calif. I Litton Industries, Inc., Beverly
Hills, Calif. I A digital-to-analogul
converter.

INDEX-

Following is the index of advertisements. Each item contains:
Name and address of the advertiser I page number where the
advertisement appears / name of agency if any.

Philco Corp., Government & Industrial Group, Computer
Div., 3900 Welsh Rd., Willow Grove, Pa. I Pages 12,
13 I Maxwell As: xiates, Inc.

c. P. Clare & Co.,

3101 Pratt Blvd., Chicago 45, Ill. / Page
& Finn

Radio Corp. of At: :::rica, Electronic Data Processing Div.,
Camden 2, N.J. I Pages 31,32 I AI Paul Lefton Co., Inc.

I Page 27

Reeves Soundcraft Corp., Great Pasture Rd., Danbury,
Conn . .I Page 9 / The Wexton Co., Inc.

23 I Reincke, Meyer

Clary Corp., Computer Div., San Gabriel, Calif.
I Erwin Wasey, Ruthrauff & Ryan, Inc.

General Electric Co., 3198 Chestnut St., Philadelphia 4, Pa.
I Page 17 I Deutsch & Shea, Inc.
Minneapolis Honeywell, Datamatic Div., Wellesley Hills
81, Mass. I Pages 6, 7 I Batten, Barton, Durstine & Osborn
Philco Corp., Computer Div., Willow Grove, Pa.
I Maxwell Associates, Inc.
30

I

Page 3

Royal McBee Corp., Data Processing Div., Port Chester,
N.Y. I Page 2 I c. J. LaRoche & Co., Inc.
System Development Corp., 2406 Colorado Ave., Santa
Monica, Calif. I Page 5 / Fuller & Smith & Ross, Inc.
Technical Operations, Inc., 3520 Prospect St., N.W., Washington 7, D.C. I Page 28 I Dawson Macleod & Stivers
Technical Operations, Inc., 305 Webster St., Monterey,
Calif. I Page 29 / Daws~n Macleod & Stivers
COMPUTERS and AUTOMATION for May, 1960

2,922,143 I Herman Epstein, Philadelphia,

Pa. I Burroughs Corp., Detroit, Mich.

I A binary storage system.
2,922;144 I Donald F. White, Oakland,

and William H. Reinholtz, Berkeley,
Calif. I Smith-Corona Marchant, Inc.,
a Corp. of New York I A read-record
circuit.
2,922,145 I Andrew H. Bobeck, Chatham,
N.J. I Bell Telephone Lab., Inc., New
York1 N.Y. I A magnetic core switching
circuit.
January 26, 1960: 2,922,231 I Victor R.
Witt, Poughkeepsie, and Rex C. Bradford, Wappinger Falls, N.Y. I I.B.M.
Corp., New York, N.Y. I A recording
and reading transducer for permitting
simultaneous recording and reading in
closely spaced storage areas of a traveling magnetizable record medium.
2,922,577 I Luciano Cignetti and Siegfried Reisch, Ivrea, Italy lIng. C. Olivetti & C, S.p.A., Ivrea, Italy I A digital
computing apparatus.
2,922,578 I Gareth M. Davidson, Bronx,
N.Y. I American Bosch Arma Corp.,
a Corp. of N.Y. I A trigonometric con
verter.
2,922,579 I David A. Goldman, Yorktown Heights, N.Y. I - - I An electromechanical vector solver and resolver
computer.
2,922,900 I Carl K. Gieringer, Cincinnat~,
Ohio I The Cincinnati Time Recorde;:
Co., Cincinnati, Ohio I A program
timer.
2.922,983 I Arthur H. Dickinson, Greenwich, Conn. I I.B.M. Corp., New York,
N.Y. I A data processing machine.
2,922,985 I David]. Crawford, Poughkeepsie, N.Y. I I.B.M. Corp., New
York, N.Y. I A shifting register and
storage device therefor.
2,922,986 I Alan G. Chynoweth, New
Providence, N.]. I Bell Telephone Lab.,
Inc., New York, N.Y. I A ferroelectric
memory device.
2,922,9.87 I George Haugk, Succasunna.
N.]. I Bell Telephone Lab., Inc., New
York, N.Y. I An information storage
system.
2,922,988 I Charles W. Rosenthal, New

York, N.Y. I Bell Telephone Lab., Inc.,
New York, N.Y. I A magnetic core
memory circuit.
2,922,989 I Myron ]. Mendelson and
Marc Shiowitz, Los Angeles, Calif. I
The National Cash Register Co., Day.
ton, Ohio I A computer input data control system.
2,922,990 I Ralph A. Anderson, Deerfield, Ill. I Information Systems, Inc.,
Skokie, Ill. I A data reduction system.
2,922,991 I Robert L. Frank, Great Neck,
N.Y. I Sperry Rand Corp., a Corp. of
Del. I A plural speed data receiver.
Feb. 2, 1960: 2,923,468 I Sigmund Rappaport, Port Washington, N. Y. I
Sperry Rand Corp., a corp. of Del. I
A component solving and integrating
'iystem.
2,923,469 I William W. Woodbury, San
Jose, Calif. I I.B.M. Corp., New York,
N.Y. I An electronic calculator.
COMPUTERS and AUTOMATION for May, 1960

To the systems engineer experienced in

COMMUNICATIONS:

you can help make the next big advance in

ELECTRONIC DATA PROCESSING
••• with RCA!
RCA, producer of the latest advance in automatic data
processing, the RCA 501, is now engaged in developing
concepts and equipment that will signify yet another
major breakthrough in the field. To round out our creative staff, we need a few unusually qualified systems engineers.
To qualify for one of these select positions, you should
have experience in communications techniques . . .
a familiarity with digital computer design and logical
organization of data and programming . . . and the
ability to apply advanced computer and information
theory concepts to the handling of digital data traffic
between automatic switching systems and from tributary stations.
Above all, you should have distinct ideas on how to better
the state of your art, and an abiding desire to create.

If you possess these qualifications and are ambitious for
a rewarding career with a progressive company, which
is constantly looking to the future in all fields of electronics, we strongly urge you to send a resume now to:
Mr. E. C. Baggett, Dept. CA-50
Electronic Data Processing Division
RCA, Building 10-2
Camden 2, New Jersey

. . RADIO CORPORATION
~®
of AMERICA
ELECTRONIC DATA PROCESSING DIVISION

31

BIG NEWS
in ELECTRONIC DATA PROCESSING
Two new systems, latest in a series of EDP Advancements from RCA
-world leader in electronics-join the established RCA 501
For Business in the Smaller Ranges· Divisional Ope rations

THE
NEW

RCA 301

COMPLETE, Compact EDP System at Low Cost . ..
with Optimum Work Power and Speed

• DATA STORAGE OPTION: magnetic tape or
new design discs that switch easily for changing
jobs-23 million character capacity
• HIGH VERSATILITY: paper tape, magnetic
tape, punch car d, 01' combinations, for input-output
• GREATEST CAPACITY, fastest core memory in
price cIass-20,000 character storage; 7 millionths
of a second access time; variable word length
• FULL SPEED -reads 600 cards/minu te; prints
90 0 numeric or 600 alpha numeric Jines p er minu te
• AUXILIARY ROLE-feeder or extender to 501
or 601 EDP Systems
• SIMULTANEITY OPTION permits large increase in workpower

For Large Business or Scientific Operations
TREMENDOUS WorkPower ... in Every Field of
Data Processing . . . with Previously Unattainable CAPACITY-EXPANSIBILITY-ECONOMY

• ULTRA-FAST-to 180,000 digits /se cond with
magnetic tape

THE
NEW

R

601

• EXCEPTIONAL COMPUTING SPEED, 1.5
millionths of a second "add" time, 1.5 millionths
of a second memory access t ime
• PARALLEL PROCESSING permits handling a
multiplicity of independent programs with changeable priority
• UNPRECEDENTED COMPATIBILITY with RCA data processing and communications
equipment and other processors
• OMNI-CHANNEL INPUT / OUTPUT permits
t ransfer of information to and from as many as
64 sources, with concurrent computing

For information write: Electronic Data
Processing Division, Radio Corporation of
America, Camden 2, New Jersey.

The Most Trusted Name in Electronics
RADIO CORPORATION OF AMERICA



---

Source Exif Data:

File Type                       : PDF
File Type Extension             : pdf
MIME Type                       : application/pdf
PDF Version                     : 1.3
Linearized                      : No
XMP Toolkit                     : Adobe XMP Core 4.2.1-c041 52.342996, 2008/05/07-21:37:19
Create Date                     : 2015:08:25 18:59:53-08:00
Modify Date                     : 2015:08:25 18:18:03-07:00
Metadata Date                   : 2015:08:25 18:18:03-07:00
Producer                        : Adobe Acrobat 9.0 Paper Capture Plug-in
Format                          : application/pdf
Document ID                     : uuid:911b60ee-c791-b541-a6cb-7b203ae07f6a
Instance ID                     : uuid:c350f190-d94c-b648-af27-0601280ef2e3
Page Layout                     : SinglePage
Page Mode                       : UseNone
Page Count                      : 44

EXIF Metadata provided by EXIF.tools